All operations that are applied to a clip are generated using Effects. A clip can have multiple effects applied to it, and each effect is stored in the clip's Effect Stack.
The Cinema shows the result of applying the effect stack to the clip, up to (and including) the current effect, which is highlighted in yellow in Effect Stack Manager. The current effect can be changed by pressing the Up/Down cursor keys, or by clicking on the effect in the Effect Stack Manager. As the current effect is changed using the cursor keys, the PFClean interface will update to show controls and tools that are appropriate for that effect.
All effects share some common controls shown below:
The Reset button resets the current effect's parameters to their default values.
The Disable button disables the current effect.
The Help button opens the documentation for the current effect.
The viewer tools button will toggle the viewer measurement tools. These display the colour attributes of the currently displayed frame. The sampling resolution of the instruments can be set by right-clicking on the tool button. This bring up a pop-up menu listing the available resolutions, from 1:1 to 16:1. The default sampling resolution is 4:1.
To analyse only a certain portion of the image, toggle the Area/Slice button . A right click on the button will pop up a menu allowing selection of "Slice" or "Area" mode; when disabled, the full image is processed.
When "Area" mode (default) is selected, dragging with the right mouse in the Cinema window will draw a sampling rectangle. All pixels within this rectangle are processed (according to the sampling resolution) instead of the full image. The rectangle can be adjusted by dragging the corners with the right mouse button.
When "Slice" mode is selected, dragging with the right mouse will draw the sampling slice line. When using slice mode, all pixels along the line will be processed to generate the histogram data. As with the area rectangle, the slice line can be adjusted by dragging the end points of the line using the right-mouse button in the Cinema window.
The Gamut Warning button will highlight in blue the colours which cannot be reproduced accurately on film.
Effects are created in the Effect Stack Manager as described earlier in this chapter.
Auto De-Flicker is used to remove high frequency fluctuations in image brightness from the clip. High-frequency flicker varies both temporally (i.e. from frame-to-frame) and spatially (within the image area of a single frame).
High-frequency flicker that varies both spatially and temporally is corrected by specifying a block size and processing the effect to detect the brightness changes required to remove flicker inside each block.
Channels: Specify which of the red, green and blue channels should be used in the deflicker calculations.
Range: The number of frames on either side of the current frame will be used to smooth out brightness fluctuations when correcting high-frequency flicker. The default value is 3, which corresponds to using 3 preceding frames and 3 following frames.
Block Size: The spatial resolution for correcting high-frequency flicker that varies spatially throughout each frame (the block size is measured in pixels). The image will be divided into a number of blocks and an independent flicker correction will be applied to each block. This value can be increased if flicker varies smoothly within one frame, or decreased if it varies less smoothly.
Smoothness %: Similarity of the flicker corrections for each block. A value of 0 means that the flicker corrections will be completely independent for each block. A value of 100 means the corrections for neighbouring blocks will be very similar. The default value is 75%
Red, Green and Blue Sensitivity %: The maximum amount of intensity change in each channel that should be considered as flicker. The default values are 10%, meaning that any pixel change above 10% (or roughly 102 out of 1024 intensity values for a 10-bit image) should be considered as image motion rather than flicker. If there is still significant flickering in the image after processing to detect flicker, increasing this value and processing again may improve the result. If the resulting image contains artefacts that were not present before, the sensitivity value should be reduced.
Gang RGB: When enabled, adjustments to a Sensitivity % values are made to all channels at the same time.
Luminance: When enabled, flicker is corrected in the luminance channel only, else flicker is corrected separately in the red, green and blue channels. This button should be enabled for grey-scale footage, and disabled if flicker is different in any of the colour channels.
The Auto De-Warp effect can be used to remove warps, such as squash and stretch distortions, from a clip. The main difference between the grid based De-Warp effect and Auto De-Warp is that Auto De-Warp uses motion estimation to detect warps.
Warps often occur in the first and last frames of a shot. The Handles controls can be used to limit Auto De-Warp processing to a specified number of frames at the beginning and end of a clip. Handles are turned off by default and can be activated by clicking the button in the Handles panel.
Proxy: Select a lower resolution proxy to calculate motion estimation. A lower resolution proxy will mean that motion estimation is performed faster than normal, but at the cost of decreased accuracy.
Severity: The complexity of the warp in the clip. Setting this parameter to High will require higher accuracy to calculate the fix and will increase processing time.
Range: The number of frames on either side of the current frame used to detect and fix warps in the clip. The default value is 5, which corresponds to using 5 preceding frames and 5 following frames.
Min Conf. %: The minimum quality confidence of detection for each frame. When detecting warps in the clip, Auto De-Warp calculates a measurement of quality of the detection for each frame. The default value of 50 means that only frames that meet a detection quality of 50% or more will be fixed by Auto De-Warp.
Auto Dirt Fix
The Auto Dirt Fix effect can be used to automatically remove specks of dirt, dust etc from a clip. Dirt areas are identified by examining the differences between multiple frames of the clip, and can be cleaned using either a temporal or spatial filling algorithm. Dirt areas can also be identified using a pre-generated defect-map that is encoded in the clip's alpha channel, as produced by infra red scanning technology.
These tools are intended to detect and fix artefacts that appear in a single frame only, such as spots of dirt or dust.
The Auto Dirt Fix effect will clean non-persistent dirt that appears for a single frame, unless the lookahead parameter is changed.
The motion analysis box contains parameters that affect the accuracy and speed of the motion analysis required for dirt detection or temporal cleanup.
Accuracy: The accuracy of motion estimation when performing dirt detection and temporal cleanup. Choices are Low, Normal and High. Setting this to Low will mean that motion estimation is performed faster than Normal, but at the cost of decreased accuracy.
Lookahead: The number of frames to search behind and in front of the current frame when detecting and fixing dirt pixels. By default this is set to 1, which means that pixels from the previous frame and the next frame will be used. With a lookahead value of 1, dirt can only be detected if it persists for a single frame.
Smoothness: The overall smoothness of the motion field to be controlled. Increasing this value will mean that smoother motion fields are generated, possibly at the expense of decreased accuracy around edges of moving objects.
Detection and Fix Parameters
The Detection and Fix boxes contain the parameters that are used to detect and fix dirt areas respectively.
Type: The type of dirt that will be detected (Dark and/or Light). Dark and Light refer to the relative brightness of dirt pixels against the background.
Min Contrast (%): The minimum contrast that a piece of dirt must have against the background. Here, contrast refers to the difference in pixel intensity in the range from 0% to 100%. Larger contrast values will mean that only dirt areas that look very different from the background will be detected. The default value is 10%.
Size: The minimum and maximum number of pixels a piece of dirt can contain. By default, the minimum size is 10 pixels (at 2K image resolution), and the maximum size edit box is empty, which means there is no upper limit in place. Type a number in the box to adjust either the minimum and/or maximum size (or click and drag with the left mouse button to increase/decrease the values). To remove a constraint, just delete all text from either of the edit boxes and press the Return key.
Motion: Minimum and maximum; restrict cleanup to parts of the image which undergo a range of motion. For example, by setting a maximum motion value, cleanup can be restricted to parts of the image that are slowly moving. The minimum and maximum motion edit boxes function in a similar way to the minimum and maximum size parameters. Clearing a value from an edit box will remove that constraint, so setting the minimum motion to 10 and clearing the maximum motion box will mean only pieces of dirt that move at least 10 pixels between frames will be cleaned.
Channels: The R/G/B channels be used to detect dirt. In some cases, you may get better results by detecting dirt in fewer channels but fixing in all channels. For example, if the blue channel contains a significant amount of noise compared to red and green, it will be easier to identify dirt areas if the blue channel is ignored during the detection process.
Use Defect Map: The data source used to find dirt pixels. If the clip contains a pre-generated defect map the button becomes active. When enabled, the defect map will be used to detect dirt. With the button greyed out or not pressed, image pixels will be used to identify dirt (see the Defect Map section below).
Sample: Estimate suitable parameters by analysing an area of the image. With Sample enabled draw a rectangle in the Cinema over a piece of dirt. Click and drag with the left mouse button held to draw the rectangle. Upon releasing the mouse button, parameters suitable to fix that piece of dirt will be set in the interface.
Type: Spa. (Spatial) and Tem. (Temporal) buttons specify the method used to fix dirt pixels after they have been identified. Temporal fixes will use pixels from nearby frames, and Spatial fixes will use nearby pixels from the same frame. If both, Spatial and Temporal are selected, PFClean will use the Threshold parameter to switch automatically between spatial and temporal fixes based on the size of the fix. Pieces of dirt that are larger than this threshold will be fixed temporally, while dirt smaller than the threshold will be fixed spatially. The Motion button, located next to the Threshold parameter, specifies whether the threshold corresponds to area size (when switch off) or area motion (pressed). When switched on, the decision to clean an area spatially or temporally is made according to how far that area moves between frames. Areas above the threshold are cleaned spatially, and areas below the threshold (i.e. Those that are moving slowly) will be cleaned temporally. This can be used as a quick alternative to using multiple Auto Dirt Fixes with different the minimum and maximum motion parameters described above.
The L button links Temporal and Spatial fix modes. It can be used to control what happens to temporally fixed dirt areas when the fix does not match the image data well. This can happen in situations where either there is significant variation between the colour balance in neighbouring frames, or when the motion in the image is such that motion analysis cannot accurately estimate where pixels are moving.
By default this option is switched off, which means that temporally fixed areas that do not match the image are rejected. Switching this option on will mean that these areas are fixed spatially instead of temporally. This means that these areas will not be rejected, but may introduce additional false-positives, where parts of the image are identified as dirty and cleaned spatially when in fact there is no dirt present.
D (De-flicker): When enabled scale illumination differences between nearby frames will be corrected when fixing temporally cleaned areas.
Dilation Size: The number of pixels each dirt area will be expanded horizontally and vertically by before fixing. This can be used to increase the size of each dirt area slightly and improve the quality of the fix. Unless a defect map is being used, the default value is 1, indicating dirt areas will be expanded by one pixel in each direction. When the G button is enabled both horizontal and vertical dilations will be adjusted at the same time.
Blend Radius: When enabled, detected dirt areas are expanded and blended with the background image after cleaning, reducing visible artefacts around the edges of each area.
Channels: The R/G/B channels that will be fixed.
Grain/Noise Preset: The grain preset to render grain or noise over the top of each fix. Rendering grain over the top of a fix will often improve overall appearance.
Store Fix Results To Disk: When enabled, fixed dirt areas are written to disk as they are generated. Enabling this option will increase overall performance when moving around the clip, because dirt fixes do not need to be re-calculated when returning to a previous frame. However, this will also increase the amount of disk storage required to hold the project files. Switching this option off will delete all disk files containing the areas that the effect has stored. The default state of this option can be controlled using a preference available in the General Preferences menu.
The report graph shows how much work has been done for each frame.
The report can display the results as either fixed pixels per frame (Pix), percent of image data per frame (%) or fixed areas per frame (Fix).
Please note that report data is only available for frames that have been processed, either by displaying them in the Cinema or by caching the clip.
Show: Display an overlay in the Cinema window indicating which pixels have been fixed. This can be used to check that the current parameters are suitable for detecting defects in the clip. If the overlay is visible whilst parameters are being adjusted, the preview will update automatically. To change the colour of the overlay, right-click on the button and use the colour editor to generate a new colour.
Defect maps come into PFClean via the alpha channel of the clip, and indicate which pixels need cleaning using information generated by a film scanner during the scanning process. The controls for enabling a defect map are in the Clip Panel, show below:
If the clip has a defect map stored in the alpha channel, the first thing to do is to change the Defect Map parameter in the Clip panel to either White Alpha or Black Alpha depending on whether dirt pixels are stored as alpha= 1 (white) or alpha= 0 (black). Once this is set, the defect map information will be stored and the pixels removed from the alpha channel. The View and Defect Threshold controls can be used to ensure that the pixels are correctly classified as defects.
To clean dirt using information provided by a defect map, first create a new Auto Dirt Fix effect. Provided that the Defect Map clip control has been set to either White Alpha or Black Alpha, Use Defect Map will automatically be selected, and the Fix Mode will default to Spa.+Tem.
Spatial fixing is suitable for most small defects and has the benefit of not requiring any motion analysis. Because of this, spatial correction of a defect map can be performed very quickly. In certain cases, using a Temporal fix will give improved results, especially if there are large defects in the image.
The Auto Stabilize effect is used to remove high frequency fluctuations in frame position throughout the clip.
Range: The number of frames on either side of the current frame to use to smooth out motion fluctuations. The default value is 5, which corresponds to averaging using 5 preceding frames and 5 following frames.
Max Offset X and Max Offset Y: The maximum correction (in pixels) in horizontal and vertical direction applied by the effect.
The Blocking Reduction effect can be used to reduce image artefactsthat occur when an image is encoded using lossy block-basedcompression such as JPEG (or similar DCT- based algorithms).
Edge Removal: The minimum pixel difference that must occur across the edge of a block for that edge to be smoothed. Increasing this parameter will smooth more blocks, but may also reduce detail in the image.
Spike Removal: The removal of very light or dark pixels that can often occur at the corners and along the edges of each block after edges have been removed. By default, this value is zero. The parameter corresponds to the minimum pixel difference between a spike and its immediate neighbours. Increasing this parameter will mean that more spikes are removed, but may also reduce detail in the image. The example images below show blocking reduction with and without spike removal:
Block Smoothing: Smooth out areas where compression is so high that entire areas of the image have been replaced by relatively flat blocks. The example images below show blocking reduction with and without block smoothing:
Block Size: The resolution of the blocks in the luma and chroma components. Using JPEG as an example again, the chroma information is stored at half the resolution of the luma information, so with luma information encoded in 8x8 blocks, and chroma information in 16x16 blocks, the block size menu would be set to 8:16:16. Alternatively, 8:8:8 can also be specified, for cases where the chroma blocks are also 8x8 in each channel.
Block Alignment: The location of a complete compression block. The default value is Top + Left, which will suffice in situations where the top-left corner of the image contains a complete block. If the image has been cropped, so that the top-left corner only contains a partial block, then the block alignment should be changed to identify a corner where a complete block is visible.
Colour Space: The colour space that the image was encoded with (choices are RGB, YUV, YIQ and YCbCr). Lossy compression algorithms such as JPEG encode the image in a series of blocks, often using blocks of different sizes in the chroma and luma components of the image. For example, the JPEG compression algorithm often uses the YCbCr colour space.
The Blur effect applies a Gaussian blur to the clip.
There are separate controls for the horizontal (X) and vertical (Y) Size of the Gaussian filter in pixels, in the red, green and blue channels.
The Gang RGB and Gang XY buttons will allow adjustment of all RGB channels at the same time, or all X/ Y values at the same time. Blur filter sizes are available for editing in the F-Curve editor and Dope Sheet.
The Constant checkbox can be used to ensure the blur parameters remain constant throughout the clip. When this box is not ticked, blur parameters can be edited for individual frames.
The Channel Align effect can be used to re-align the Red/Green/Blue channels of a clip from tri-strip or consecutive archive footage.
The process is automatic and uses motion estimation to calculate the correct relative pixel position within each colour channel. While the process is automatic there are a number of user definable parameters, these are detailed below.
The Channel Align effect will calculate the necessary alignment for each frame automatically. The Align Frame button can be used to re-calculate the alignment for the current frame, potentially overwriting manual adjustments.
The Translation edit boxes show the transformation values for the current frame. The values can be manually edited to fine tune the alignment.
Motion Model: The type of motion that will be used to re-align the channels. The available options are Translation, Horizontal or Vertical which correspond to alignment in both the X and Y directions, and X and Y only.
Align To Channel: The channel to align to. For example, if this is set to Red then the Green and/or Blue channels will be aligned to match the Red channel, which itself will not be adjusted.
Adjust Channels: The channels which will be adjusted to align them with the channel specified in Align to Channel.
Constant: If enabled, the alignment is constant over the entire clip, When this is un-checked, each frame may have a separate alignment transformation.
Align Clip: Start the alignment process for the whole clip. Once finished, the transformation values are available in the Transformation edit boxes, F-Curve editor and Dope-Sheet for further editing.
Note: The number of CPUs that are used to perform channel alignment is specified using the Use CPUs For Processing option in the General section of the Preferences window. Processing high resolution clips on 32-bit systems using large numbers of CPU cores may result in an out-of-memory error. If this occurs, the amount of memory used during channel alignment can be reduced by lowering the number of CPUs that are used for processing.*
The Channel Matrix effect can be used to re-map colour channels.
To re-map colours from any colour channel to another in the Rewire box, click a channel on the left with the left mouse button and drag it to the required value on the right, releasing the mouse button to re-wire the connections. Clicking on an already existing connection deletes the connection.
The Matrix corrector can be used to create a numerical mapping between channels. This could be used to create a Black and White image from an RGB colour for example.
The Chroma Cleanup effect can be used to reduce colour fringing in video footage caused by cross-talk between the luma and chroma video signals, and operates in YUV colour space. See below for a before and after screenshot illustrating the type of colour fringing this effect is designed to reduce.
Luma Threshold %: The threshold used to identify bright areas in the image around which colour fringing occurs. Decreasing this value will remove colour fringes from dark and bright areas, and increasing this value will remove colour fringes only from bright areas. The default value is 20%.
Luma Similarity %: The similarity of nearby pixels to be used to calculate the cleaned chroma values. Decreasing this value will mean a wider range of pixels are used to calculate the cleaned chroma value, but at the cost of potentially introducing some colour flattening. The default value is 80%.
Chroma Threshold %: The threshold to determine which pixels are considered to be colour fringes that require cleaning. Pixels with chroma values above this threshold are assumed to be corrupted when applying the filter. The default value is 40%.
Filter Width and Filter Height: The size of the area used to calculate the cleaned chroma values. The default value is 4 for each parameter.
Constant: When checked, the parameters should be held constant over the entire clip or allowed to vary. When parameters vary, their F-Curves are available for editing in the F-Curve editor and Dope Sheet.
The image below shows before and after results when using the Chroma Cleanup effect:
The Chroma Re-Sampling effect can be used to correct artefacts often found in video footage that occur when the chroma information in the image is encoded at a lower resolution than the luma information.
The sampling scheme is specified using a fixed number of luma and chroma samples, such as 4:1:1 which indicates one chroma sample (in two channels) is used for every 4 luma samples.
After specifying the sampling scheme using the Chroma Type menu, or by choosing a preset from the Preset menu, the resolution of the chroma information in the image will be increased automatically, to match the resolution of the luma information (also known as 4:4:4, indicating there are 4 chroma samples in two channels for every 4 luma samples). Chroma types currently supported are 4:1:1, 4:2:0, 4:2:2 and 3:1:1. The Colour Space menu can be used to specify the format in which chroma was originally encoded in the clip.
Presets for various common video formats are listed in the Preset menu, and choosing a preset will automatically set the Colour Space and Chroma Type menus to the correct values. Presets are available for DV (PAL and NTSC), DVCPRO (PAL), DVCPRO50, DVCPRO HD, Digital Betacam, Betacam SX, MPEG IMX, HDCAM, HDCAM SR and HDV.
The Smoothing parameter can be used to control how much colour information is smoothed across edges in the image, with a default value is 10%. The Constant tick-box specifies whether the smoothing parameter is constant throughout the clip or not. If not, the smoothing parameter is available for editing in the F-Curve editor and Dope Sheet.
An example of chroma re-sampling is shown below, where an image encoded as 4:1:1 (on the right) is re-sampled back to 4:4:4 (on the left). The artefacts caused by the low resolution chroma information are no longer visible in the left-hand image.
Clean Lens Dirt
The Clean Lens Dirt effect can be used to remove dirt that persists in the same place from frame-to- frame. Typically, this is caused by marks on the camera lens, rather than by specs of dirt on the original film.
The Clean Lens Dirt effect will fix persistent dirt that stays fixed in the same place in multiple frames. For non-persistent dirt that is typically present in only one frame, the Auto Dirt Fix effect should be used.
The Clean Lens Dirt controls are shown below:
Frame Range: The frames that will be cleaned. This can be set to Clip to process the entire clip, or From/To to process a specific range of frames.
From, To: The start and end frame to be cleaned when the frame range is set to From/To. Clicking on one of the buttons will put the current frame number in the edit box.
Type: The type of dirt that will be detected (Dark Only, Light Only, or Dark and Light). Dark and Light refer to the relative brightness of dirt pixels against the background.
Channels: The red, green and blue channels will be used to detect and fix lens dirt.
Mark Area: Allow a rectangle to be drawn around each piece of lens dirt that requires fixing. Because lens dirt is persistent from frame-to-frame, there is no need to do this in every frame. Marking a piece of dirt in one frame is sufficient for it to be fixed in all frames where it is detected. To mark an area, click the button and then draw a rectangle in the Cinema around the piece of dirt by clicking and dragging with the left mouse button held. After an area is marked, it appears in the lens dirt list, described below.
Start: Start the process of cleaning lens dirt.
Lens Dirt List
The lens dirt list contains all pieces of dirt that have been marked in the clip.
A piece of dirt can be renamed by clicking in the Name column and entering a new name. The Significance column shows the total pixel intensity change. Pieces of dirt with larger significance values are generally more visible than those with smaller significance. The list can be sorted by clicking on any of the headers. To select a specific piece of dirt, click on a row in the dirt list with the left mouse button. If the Highlight button is pressed, selected pieces of dirt will be highlighted in the Cinema. To select multiple pieces at the same time, hold the Shift or Control keys whilst clicking.
After one or more pieces of dirt have been selected, clicking the Enable or Disable buttons will enable or disable those pieces. The All/None button under the dirt list can be used to select all or none of the pieces in the list at the same time. Clicking the Delete button will delete all selected pieces. The Show Selected button will shorten the list so only the selected pieces are visible. Clicking the button again will revert back to showing all pieces of dirt. The Marquee button is used to draw a rectangle in the Cinema and select all pieces of dirt within that rectangle. Clicking the Highlight button will toggle highlights on/off in the Cinema, and the Names will display a name as well as a highlight.
Clear Fixes: Clear the fixes of any lens dirt area that has been marked. Once fixes have been cleared for an area, they can be regenerated by clicking the Start button again.
Undo and Redo: Undo/redo the last edits applied to the effect. The number of undo levels can be changed in the Preferences window.
The Colour Balance effect can be used to remove dominant colour tints or tones from a clip. It is important that the clip's colour space is set correctly before creating a new colour balance effect. The Colour Balance effect will automatically perform an initial operation, which can be adjusted using the controls described below.
Colour balance is mainly controlled by the Black-Point, Balance-Point and White-Point values. Clicking these buttons allows colours to be sampled from the Cinema window by clicking with the left mouse button, or by clicking and dragging to draw a sample rectangle.
When balancing a clip, the overall colour of the clip will be changed so the Black-Point is mapped to the Black-Target, the White-Point is mapped to the White-Target and the Balance-Point values is mapped to the Balance-Target.
In this way, if a clip contains a green tint for example, this tint can be removed when then Balance-Point (a green colour) is mapped to the Balance-Target (grey, or another choice of colour).
This simple colour balance approach allows footage that appears to be significantly colour-damaged to be quickly resurrected.
The drop-down menus on the right of the Black-Point, Balance-Point and White-Point parameters can be used to control which aspects of each colour are sampled (see below). The default value is Chrominance, which indicates that only the chrominance component of pixels will be sampled. Alternative values are RGB, to sample full RGB colours, or Luminance which will only sample the luminance of pixels. These menu options are useful in situations where you only want to remove a colour tint, without changing the overall brightness of the image.
The Source Colours and Target Colours tabs contain colour wheels that illustrate the colour and luminance of each parameter.
The values can be adjusted by clicking and dragging the indicator in the centre of each wheel. The luminance bars along the bottom control the overall luminance of the balance point. Clicking on the colour buttons underneath each bar will display a colour chooser where you can select pre-defined colours.
The RGB Curves tab show the RGB look-up tables that are used to balance the clip.
Spline points can be moved by clicking and dragging with the left mouse button.
The Offsets tab shows the numerical values of the look-up tables that are used to balance the clip
The Gang buttons next to each parameter can be activated so when values in the edit boxes for one channel are changed, the same change is applied to the other RGB channels. The Reset buttons will reset a colour back to its default value.
Clicking one of the Lock buttons will lock the current value so it cannot be adjusted. This is especially useful during automatic colour balance because it allows certain points (such as Balance-Point) to be estimated whilst keeping others (Black and White, for example) locked at their default values.
Automatic Colour Balance
Clicking the Balance Clip button will estimate the Black Point, Balance Point and White Point for all frames in one pass, unless they have been locked by clicking the Lock buttons in the Offsets tab. If the Constant button is also checked, these values will be averaged over the entire clip to produce a single colour balance, otherwise separate values will be stored for each frame. All parameter values are available in the F-Curve editor and Dope-Sheet for further editing.
Clicking the Balance Frame button will estimate the colour balance for the current frame. This may overwrite any manual adjustment done previously. If the Constant button is checked, these values will be used throughout the clip.
Manual Colour Balance
Manual colour balance can be performed by clicking one of the Black Point, Balance Point or White Point buttons.
Clicking (or clicking and dragging) with the left mouse button in the Cinema window will sample a colour from the image. The edit boxes, colour wheels and channel transfer will update accordingly and the colour balance will be applied immediately. If the result is not acceptable you can pick another colour easily from another part of the image. If you are removing a colour tint, try to pick a Balance Point colour from an area of the image you know should be grey (or another known colour).
The actual values sampled from the image depend on the state of the drop-down menus on the right of each parameter. For example, when sampling the Balance Point, the default menu option of Chrominance means that only the chrominance component of colour is sampled from the image, and luminance is maintained at the same level as the Balance Target.
The Colour Transfer effect can be used to adjust the colour in a clip to better match an example image provided by the user. Source and destination areas can be specified to control how colour is transferred between the images.
To specify which clip to use as the source clip, drag a clip out of the Media Bins and drop it into the Source Clip window of the Colour Transfer panel.
Note: When transferring colour, more reliable results will be obtained if the source image is of similar composition to the working clip. Also, it is not possible to transfer colour to a monochrome clip - both the source and working clips must already contain colour variations.
Drawing Transfer Areas
In situations where the initial transfer is not satisfactory, transfer areas can be placed in the source and working clip to control the colour transfer more precisely.
Each area is specified by drawing a rectangle in both clips, and pixel colours from the source clip rectangle will be transferred to colours in the working clip rectangle.
Each area must be placed in both the source clip and the working clip, so use the Show Source Clip button to display the source clip and place each area.
The example below shows two transfer areas placed in the source and working clip. The source clip is on the left, and the working clip is on the right. In this case, we wish to transfer the colour appearance from the left-hand clip onto the right-hand clip.
The first area was drawn around the crowd in the background, and the second area on a face in the foreground. The colour transfer result is shown in the middle image below, with the original clips on the left and right.
: When enabled, displays the Media Bins to select a source clip for the colour transfer.
Source Clip: Displays a thumbnail and frame number, indicating which frame of the source clip colour will be taken from. To change frame, click the left mouse button on the thumbnail and drag the mouse left/right, or enter a new frame number in the Frame edit box below.
Show: Display the source clip in the Cinema window.
Remove: Remove the source clip from the effect.
New: Create a new area and add it to the list. Areas that have not yet been positioned in both clips are greyed out in the list. To place an area in a clip, click on the area with the left mouse button to select it in the list, and then click-and-drag with the left mouse button in the Cinema window to draw the rectangle. To change the position of the rectangle, click and drag again in the Cinema window.
To change the display colour for each area, double-click in the Border column to display a colour chooser. Note that the RGB colour of each area does not affect the colour transfer - it is only used to distinguish between different areas in the Cinema window. To disable an area, click in the Ignore column. Disabled areas will not be considered in the colour transfer.
Delete: Delete the selected area.
Smoothness: Smoothness of the overall transfer.
Decreasing the smoothness control can often provide sharper transfers that only affect parts of the colour space (the default value is 50%).
Transfer: Calculate the transfer coefficients and apply the transfer to the working clip in the Cinema window. It is only necessary to transfer colour once for the entire clip. If separate colour transfers are required for different parts of the clip then it is recommended to use two separate Colour Transfer effects.
The De-Flicker effect can be used to remove low frequency fluctuations in image brightness from the clip. Flicker often comprises of two separate artefacts: a high-frequency flicker that varies both temporally (i.e. from frame-to-frame) and spatially (within the image area of a single frame); and a low-freqeuency flicker that varies temporally. The De-flicker effect can be used to correct low frequency fluctuations.
Once areas have been drawn, the De-Flicker effect will analyse each frame of the clip and determine the corrections that need to be made to each user-defined area to correct flickering. Once completed, the brightness curves for each user-defined area will be available in the F-Curve editor and Dope Sheet. Note that if the ROI for the clip or effect is changed, de-flicker corrections must be re-calculated.
Note that the brightness changes for each user-defined area are still applied to pixels outside each area. When more than one area is present, the changes are smoothly interpolated between areas. Generally, it is better to place user-defined areas in parts of the image that do not contain significant motion or edges. The screenshot below shows one user-defined area that has been placed in a relatively flat region at the top-right of the image.
Add Area: Draw a rectangle in the Cinema by clicking and dragging with the left mouse button. To change the size or position of an area, click and drag with the left mouse button whilst hovering the cursor over the centre or edge of an area in the Cinema window.
Areas can be animated to change position throughout the clip if required (for example, if the camera or scene is moving). Clicking the In or Out buttons will set the in-point or out-point for the area to the current frame. The range of frames between the in and out-point for the currently selected area is shown in the scrub bar with a blue highlight.
Keyframes are shown in the scrub bar as vertical yellow markers. Keyframes are generated automatically when creating or editing areas in the Cinema window. Clicking the K+ or K- buttons will either add or remove a keyframe at the current frame, and the Prev and Next buttons can be used to move to the previous or next keyframe.
Marquee: Select areas from the Cinema window. After clicking the button, draw a rectangle with the left mouse button to select areas (selected areas are displayed in yellow). If multiple areas are selected whilst one is edited, their size and position will all be adjusted at the same time.
All/None: Select all areas or unselect all areas, depending on how many areas are selected when the button is clicked.
Delete Areas: Delete all selected areas.
Filter Type: The type of filter used to smooth out intensity variations, either Median, Gaussian or Lock. Median and Gaussian filters are applied over the current Frame Window. The Lock filter (the default) will attempt to lock all brightness fluctuations relative to the Target Frame, rather than just smoothing them using a Gaussian or Median filter.
When using the Lock filter, the target frame can be set by clicking the Target Frame button or entering a frame number in the edit box. Clicking the button will enter the current frame number in the edit box. By default the target frame is set to the middle of the clip.
Motion: When selected, amotion correction will be attempted before estimating the amount of flicker in each area. Generally, this is only necessary when the area has not been animated fully throughout the clip, or when the area contains significant image texture or edges. This also increases the time required to process. For areas placed over relatively flat image content, this option is not often needed.
Each area is listed in the Areas table. The Show Areas button can be used to toggle display of the area boundaries in the Cinema window. Areas can be selected by clicking in the list, and re-named by double clicking in the Name column. When trackers have been activated in the De-Flicker effect, right-clicking in the Tracker column of the list will display a popup menu listing all active trackers. Attaching an area to a tracker will override any keyframe animation and instead use the tracker motion to adjust the area's position in each frame. Note than when creating trackers in the Tracker panel, it is wise to switch on the De-Flicker option when tracking. The Show Trackers button can be used to show or hide all trackers in the Cinema window.
The De-Fog effect can be used to reduce low-to-moderate levels of fogging caused by X-Ray exposure or similar exposures. Typically, fogging is seen as rapidly moving bands of colour throughout the clip, corresponding to the parts of each frame that have been exposed to the fogging source.
For fogging to be removed successfully, it must not be so strong that it destroys the underlying image data. Only low-to-moderate levels of fog can be removed automatically using this effect. Typically, after exposure to fogging, the blue channel is the most severely affected, and dark areas of the image are affected more than light areas.
Window: The number of other frames that are examined on either side of the current frame, when trying to detect which parts of the image are fogged. The default value is 1 which means that the current frame will be compared to the previous and next frames. If fog effects are slowly moving through the clip, this value should be increased to ensure that pixels can be compared against those in another frame that does not contain any fogging artifacts.
Red Fog %, Green Fog % and Blue Fog %: The amount of fog that is present in the clip. For example, the default blue fog value is 30% which indicates that when exposed to the source of fogging, the maximum intensity change undergone by any pixel is 30%, corresponding to roughly 300 our of 1024 intensity values for a 10-bit image. It is important that the fog levels entered in the UI correspond approximately to the actual fog levels in the image. Most often, this means that the blue channel has the largest value, followed by the green and then the red. Checking the Gang RGB button will ensure that these levels are adjusted together.
Show: Display an overlay showing which pixels have been adjusted during fog removal.
Sample: Sample the image to obtain a rough estimate of the amount of fog in an image. Use the left mouse button to draw a sample rectangle in the Cinema. Upon releasing the mouse button, the fog level in each channel will update. For best results, sampling should be done in a dark area of the image, where the fogging is most visible.
Constant: When selected, the fog levels are constant throughout the clip or varying. This can be used in situations where some frames exhibit more fogging than others. The fog levels are also available for editing in the F-Curve Editor and Dope Sheet.
The De-Grain effect uses grain presets to remove grain from the clip.
Mode: De-grain each channel separately (RGB) in each of the red, green and blue channels (RGB), or all channels of the image are de-grained together (Grey).
Temporal: When selected, the De-Grain effect uses a temporal degrain algorithm to remove grain. When this button is selected, multiple frames will be used simultaneously to remove grain.
Size: The expected size of grain particles, measured in pixels.
Intensity: The intensity of the grain in each channel, measured in the range [0..100].
G (Gang): Apply edits to all R/G/B channel values at the same time when selected.
Sample: Sample grain from part of the image by drawing a rectangle in the Cinema with the left mouse button. After the button is released, grain parameters will be estimated by analysing the image pixels inside the rectangle. To sample from different areas of the image at the same time, hold the Shift key whilst drawing a rectangle.
Grain analysis performs best when the rectangles are placed in relatively flat areas with a constant overall colour. After grain is analysed, the following parameters will update:
The Size text boxes show the estimated average size of the grain particles (in units of pixels).
The Intensity text boxes contain the estimated grain intensity (as a percentage in the range 0..100)
Note that grain sampling will only provide an initial estimate of the grain parameters. Grain sampling will also not function accurately for very degraded or non-uniform image data.
Low, Mid and High: Increases or decreases the amount of de-grain for low, mid and high intensity values in the image. The regions for low, mid and high can be adjusted using the pivots in the profile graph.
Level: The amount of grain removed from the clip. The default value is 1.0 and corresponds to removing all grain (as defined by the parameters) from the clip.
Edge Preservation: The amount of edge preservation. A higher value ensures that significant edges in each image are not adversely affected by grain removal.
Defringe: The amount of high frequency colour fringing to remove. Typically, such fringing occurs when large amounts of grain is present on strong edges in the image. Setting this option to 1.0 will remove more fringing than lower values.
Despeckle: The amount of noise reduction in smooth regions after grain has been removed. This noise may include artefacts such as occasional bright or dark pixels that are not removed during de-grain. Setting this option to 1.0 will remove more noise than lower values. For better control of noise, additional effects such as Noise Removal and Median Filter can be applied after de-grain.
Sharpen: Amount of sharpening applied to the image after de-graining the image. Setting this option to 1.0 will result in more sharpening than lower values. For better control of sharpening, an additional Sharpen effect can be applied after de-grain.
Grain and noise presets can be stored as XML files for use elsewhere in the application or in later projects. The Presets palette shows the presets that are currently available.
To copy the current parameters into a new preset, click the Store button. If an existing preset is highlighted when the Store button is clicked, that preset can be overwritten if desired.
To select a preset, click it with the left mouse button. Click again on the preset name to change its name. To use the parameters from one preset in the current de-grain, select the preset and click the Use button. Pressing the Delete button will delete the selected preset.
To examine the contents of each preset, hover the mouse pointer over a preset name. This will display a small popup window showing the contents of the preset.
When presets are stored, they are saved as XML text files in
$HOME/.pfclean/presets/grain/ and also recorded in the current clip.
The De-Noise effect uses noise presets to remove noise from the clip.
Intensity: The amount of noise that is removed in each of the red, green and blue channels.
Impulse: The threshold for detecting impulse noise. Impulse noise differs from normal noise in that it will appear in the clip as occasional very bright or very dark pixels of different colours. The Impulse edit boxes indicate the minimum pixel intensity difference between a pixel and its surrounds before the pixel is identified as noisy.
A value of 100% (the default) means that no impulse noise will be removed. Reducing the value to 50% would mean that impulse noise must differ from its surrounds by more than 50% of the maximum image intensity to be removed.
Note that as the impulse noise values are reduced towards zero, the time taken to detect and fix impulse noise can increase significantly.*
Mode: De-noise each channel separately (RGB) in each of the red, green and blue channels (RGB), or all channels of the image together (Grey).
Gang: When selected, all edits to any R/G/B channel values are applied to all channels at the same time.
Temporal: When selected, a temporal denoise algorithm is used to remove noise. When this button is selected, multiple frames will be used simultaneously to remove noise.
Noise Level: The amount of noise removed from the clip. Menu options are High, Medium and Low. The default value is High and corresponds to removing all noise (as defined by the parameters) from the clip.
Edge Preservation: The amount of edge preservation to ensure that significant edges in each image are not adversely affected by noise removal. The menu options are Low, Medium, High and None (the default).
The Profile button in either the De-Noise or Re-Noise effect will open a noise profile editor to allow the noise response to be adjusted for different pixel intensities.
Editing the profile graph allows adjustments to noise intensity in relation to pixel luminance values:
The horizontal axis corresponds to pixel value, and the vertical axis to intensity response.
By default, the profile is a horizontal line, which means that dark and light pixels use the same amount of noise, corresponding exactly to the parameters. The response can be adjusted so that darker or lighter pixels can use more or less noise as required. Drag the control vertices around with the left mouse button to edit the curve.
To insert a new vertex, click on the curve with the left mouse button. To make a vertex linear, drag the Bezier handles into the vertex, and to expand the handles out of a linear vertex, hold the Shift key and drag with the left mouse button. To remove a vertex, click on it with the right mouse button.
The Reset button can be used to reset the profile to its default state. The Channel menu allows the profile to be adjusted separately for each colour channel if required.
The De-Warp effect can be used to remove warps, such as squash and stretch distortions, from a clip. The De-Warp can use grids and/or individual warper points and uses our state of the art tracking system to provide an unrivalled automatic De-Warper.
The De-Warp effect works using two grids that control the warping of the clip: the Source Grid is placed on the original warped clip and edited so that its vertices are in useful positions that can subsequently be tracked or identified easily from frame-to-frame. The Destination Grid specifies where the vertices will move to, and hence control the way in which the original clip is de-warped. The Source Grid and Destination Grid can be copied and edited individually. The Source Grid can also be tracked throughout the clip to reduce the burden of placing grid vertices individually.
The most common use for the De-Warp tool is to remove vertical squash and stretch distortions in the image, for example, when the image jumps slightly up and down and squashes and stretches vertically at the same time. To correct for this type warping, the Source Grid should be tracked so the vertices follow the distortions accurately, and the Destination Grid can be left in the default state: no vertices are moved. When the source grid is warped to match the destination grid, the squash and stretch will be remove because the tracked Source Grid vertices (the left-hand image below) are mapped onto the static Destination Grid (the right-hand image below).
The Grid Controls are shown below:
Source and Destination: Display the Source Grid or Destination Grid. When the Source Grid is visible, the original clip is shown underneath. When the Destination Grid is visible, the image will be de-warped so that each Source Grid vertex is moved to the corresponding Destination Grid vertex. De-selecting both buttons will show the de-warped image without any grid overlay. The Source Grid is drawn in white, and the Destination Grid is drawn in green.
Copy Src->Dest and Copy Dest->Src: Copy the Source Grid to the Destination Grid, or vice-versa. Note: when the Source Grid and Destination Grid vertices are in the same position, no de-warp will occur.
X Vertices and Y Vertices: The number of vertices that are in the Source and Destination Grids. The number of vertices in both grids is always the same because the points always need to correspond. A higher density of vertices will mean that the grid will track and render more slowly, but will be able to represent complex warps more accurately.
Note: if you change the X or Y vertices in a grid, any tracking or keyframe information will be removed.
Warp Channels: The red/green/blue channels that will be de-warped. Any combination of channels can be used and this can be of great benefit when working with tri-strip material in which warping can occur in only a single channel.
Reset: Reset the current grid to its default position in all frames.
Reset Frame: Reset the currently selected grid to its default position in the current frame only.
The grid editing tools can be used to manually adjust the position of grid vertices.
Draw Grid: Allow a new grid to be drawn over the image. To do this, click and drag with the left mouse button in the Cinema window to draw a rectangle representing the grid boundary.
Move Grid: Adjust the spacing between grid columns and vertices. Click on a grid vertex and drag with the left mouse button to adjust the spacing and position of the grid.
Move Vertices: Allow individual vertices to be moved using the left mouse button. If more than one vertex is selected then all selected vertices will be moved together.
Move Region: Allow all vertices inside a circular region to be adjusted at the same time. To adjust the size of the region, hold the Shift key whilst clicking and dragging with the left mouse button.
Marquee: Select multiple vertices by drawing a selection rectangle in the Cinema. This is useful for adjusting their positions as a single group, or for tracking groups of vertices. Click and drag with the left mouse button in the Cinema to draw a selection rectangle. Selected vertices are drawn in light-blue. To select multiple times, hold the Shift key whilst drawing the selection rectangle.
Add Key: Add a keyframe to all selected vertices at the current frame. Frames where the current grid has been keyed are indicated in the scrub bar. Selected grid vertices are available in the F-Curve Editor and Dope-Sheet for further editing.
These controls specify the way in which grid vertices are allowed to move when they are tracked, and allow vertices in the Source Grid to be tracked from frame-to-frame.
Horizontal and Vertical: Specify how grid vertices can move in the horizontal and vertical directions. The options are No Motion to prohibit the vertices from moving in that direction, Move Vertices to allow individual motion for each vertex, Move Column and Move Row for all vertices in a single row/column moving together. Finally, Move Grid will force all vertices in the grid to move in the same way.
For example, to correct the squash and stretch situation described above where the image data is only moving vertically, Horizontal would be set to No Motion, and Vertical should be set to Move Row to indicate that all of the vertices in a single row of the grid will move together in the vertical direction.
Sensitivity: The size of the area around each grid vertex that is used to track Source Grid points. The sensitivity areas can be displayed by clicking the Show Sensitivity button in the Display part of the De-Warp interface.
<< < > >>: Track selected vertices backwards through multiple frames, backwards one frame, forwards one frame or forwards multiple frames respectively. When tracking through multiple frames, click the button again to stop tracking, or press the Escape key.
Note: only Source Grid vertices can be tracked.
Lock Unselected Vertices: When enabled, vertices that are not selected will remain locked in their current positions. When the button is not pressed, unselected vertices will have their motion estimated using nearby vertices that are being tracked. This is useful when the Source Grid extends up to the edge of the frame, which means it becomes harder to track the edge vertices accurately. By only tracking internal vertices, the motion of vertices on the edge can often be estimated more reliably.
Ignore Global Motion: When selected, global image motions (like a pan or zoom) are ignored by the tracker. When the button is pressed, only the non-global component will be tracked (i.e. Only the warp will be corrected, instead of the pan or zoom).
Warper points can be used as additional tools for warping the clip. Warper points can be positioned anywhere in the frame, and act after the Source/Destination grid warp has been applied.
To create a new warper point, switch off both the Source and Destination Grids, and use the Add Local or Add Global button to draw a warper point that has either local or global effect. To draw a warper point, click and drag with the left mouse button to place the point centre and the area of influence. Once placed, dragging the centre point will start to warp the image.
Note: when adding a new point, the image will temporarily revert back to its original state so the warper point can be placed accurately. Once placed, the warper points can be edited using the left mouse button.
Local warper points only affect the region inside the area of influence, whereas Global points affect outside the region as well. Warper points can be selected using the Marquee tool described above. Clicking the Delete Points button will delete all selected Warper Points. Warper points are available in the F-Curve and Dope-Sheet for further editing.
The Display controls affect the display of the De-Warp in the Cinema.
Render Quality: The representation of the image in the Cinema. Options are Full or a half or quarter-proxy representation of the image, which is useful for increasing interactivity whilst editing grid vertices for high resolution clips.
Show Names: Toggle display of grid vertex and warper point names.
Show Sensitivity: Toggle display of the grid vertex sensitivity areas.
Show Points: Toggle display of any warper points.
The Dustbust effect can be used to quickly remove significant non-persistent bright or dark spots from a clip.
It functions in a similar way to the Auto Dirt Fix effect but has been optimised to operate at a significantly higher speed and does not provide as many controls to the user.
The Dustbust effect performs motion compensation to be able to detect spots in parts of the image that are undergoing significant motion. This motion compensation can be performed on the full resolution image or on a smaller scale proxy.
Type: Type of spots to be removed. Options are Dark or Light and both Dark and Light can be selected simultaneously.
Detect and Fix: The possibly different channels to be used to detect and fix spots. For example, in many situations the blue channel of the clip may be corrupted by significant amounts of noise. In these cases, the accuracy of spot detection will often improve when only the red and green channels are used to detect spots.
Lookahead: The number of frames before/after the current frame used to detect spots. For a spot to be detected, it must exist in the current frame and not in the before/after frames.
Constant: When selected, the De-Spot parameters should be held constant over the entire clip or allowed to vary. When parameters vary, their F-Curves are available for editing in the F-Curve editor and Dope Sheet.
Deflicker: When selected, small scale illumination differences between nearby frames will be corrected when fixing temporally cleaned areas.
Motion: The resolution used for motion compensation. Increasing the resolution will improve the results of the motion compensation but will also increase the time it takes to apply the effect.
Thresh %: The minimum intensity difference required to detect a spot. Increasing this value will mean that only very bright (or very dark) spots are detected. Decreasing the threshold too much may result in spots being detected incorrectly.
Dilate: Amount of size increase of the detected area. Note that this size increase will be performed before any check for false positives and that increasing the size of a dust area might cause it to be rejected.
Reject %: The Dustbust effect checks each detected area and fix before applying it to the final result. The Reject parameter controls this check. Increasing this parameter will increase the likelihood of a piece of dust to be rejected as a false positive.
The Edge Framing effect can be used to remove empty borders around a stabilized frame. To use this, the clip must be prepared so that the alpha channel identifies which pixels are to be kept, and which are to be removed. Pixels to be kept must have alpha= 1, and pixels to be filled must have alpha= 0. By default, after applying a Stabilize effect any empty border areas will already have their alpha value set to 0, indicating that they are to be removed.
Mode: The method that will be used to remove pixels during the framing procedure. The default mode is Scale To Fit ROI, which indicates that the image will be scaled up to ensure that no empty borders are visible inside the current region-of-interest. When using this mode, clicking the Start button will start to search the clip to find a suitable border size. Once found, the border will be drawn as a stippled yellow rectangle in the Cinema window. This rectangle can be adjusted as required by clicking and dragging with the left mouse button. The Start button will also change label to Accept, and can be clicked again to accept the rectangle and scale the image data accordingly.
Mode can also be set to one of Translation, Affine or Perspective, which will use increasingly flexible models of motion to try and remove empty borders using pixels from adjacent frames. Translation will try to estimate only horizontal and vertical motion. Affine and Perspective will use an affine and planar-perspective transformation respectively. Masks can also be used to specify an offending area for removal. The Start button will start the edge framing processing.
Note: The number of CPUs that are used to perform edge framing is specified using the "Use CPUs For Processing" option in the General section of the Preferences window. Processing high resolution clips on 32-bit systems using large numbers of CPU cores may result in an out-of-memory error. If this occurs, the amount of memory used during edge framing can be reduced by lowering the number of CPUs that are used for processing.*
The Film Grade effect can be used to perform primary colour correction on entire frame or areas of frames. Film Grade offers a combination of 4 grading modes:
A Wheel / Trackball mode, which emulates a typical tactile control surface, and allows adjustments to Luminance (Density), Balance, Contrast and Saturation (Master, Low, Mid and High)
B Curve mode, which allows precise graphical adjustment of RGB and luminance levels (Master, Low, Mid and High).
C Histogram mode, with simple threshold and level controls. (Master only)
D Printer Light mode, which emulates a traditional optical film printer (Master only)
Master and Luma Ranges
When Master (1) is selected, grading adjustments affect all luminance levels equally. This is possible for all grading modes.
Density level and balance can be adjusted with the wheel and trackball. Slider 4 controls Contrast, and slider 5 controls Contrast Pivot (8). The Contrast Pivot is the point around which the Contrast rotates
|Contrast Pivot Default||Contrast Pivot low||Contrast Pivot high|
Slider 6 controls Saturation
When a Luma Range (Low, Mid, or High) (2) is selected, grading adjustments affects only the luminance values within the selected range. This is possible for Wheel / Trackball and Curve modes
Coarse adjustments to the luma ranges can be made by moving the Luma Range Pivots (9)
In this example, the Mids have been darkened by rotating the wheel anticlockwise:
|Mid Pivot Default||Mid Pivot low||Mid Pivot high|
Fine adjustments to the luma ranges can be made by modifying the Luma Ranges (10)
The Luma Ranges menu (10) displays editable curves used to define the range over which the Low, Mid and High tonal values operate. Values will overlap but can be adjusted to increase or decrease the tonal range represented by any of the 3 range buttons. The curves can be adjusted by clicking and dragging the curve handles with the left mouse button. New vertices can be introduced by clicking with the left mouse button on the curve. Points can be removed by clicking on them with the left mouse button
In this example, the Lows have been darkened by rotating the wheel anticlockwise: Compare the results shown in the Response Curves when the Luma Range curves are modified
Here, the Low range is narrower, so only the darkest tones are affected:
Here, the Low range is wider, so a greater range of tones, including midtones, are affected:
Curve Grading mode ( B ) will display a curve editor where the red, green, blue and luminance curves can be adjusted individually by clicking and dragging the curve handles. New vertices can be introduced by clicking with the left mouse button on the curve. Points can be removed by clicking on them with the left mouse button. CTRL_Clicking (or CMD_Clicking on OSX) will 'break' curve tangents, allowing differing input and output tension, as shown on the green channel below.
Histogram mode ( C ) opens the histogram panel. By default the complete colour range is shown but this can be adjusted using the yellow lines at the edges of the main graph window. Moving the lines will remap the black and white points to a narrower range. The Sample button is used to marquee an area of the frame to display the values within the sampled area.
Printer Light mode ( D ) emulates a traditional optical printer, allowing fixed increments to R,G,B and Luminance be entered via the GUI or numeric section of the computer keyboard
This mode can be used to recreate colour timing settings where these have been recorded in lab reports.
The Incremental Cineon code value of the printer light step (default = 12) and neutral base value (25, 25, 25) can be set in the Preferences / Grading section
The Fix Bands effect can be used to remove dark or light bands in the luma or chroma channels of video footage. This effect operates in YUV colour space, and can remove bands that are present in one field of one or more frames of the clip.
The screenshots below shows before and after results when using the Fix Bands effect for luma and chroma bands, and illustrates the type of artefact this effect has been designed to detect and fix.
Note that for video footage, these bands typically appear on only one field of each frame. Applying any sort of lossy compression to the original footage is likely to disrupt the scanlines and prevent them from being detected correctly.
When fixing luma bands, this effect can operate with interlaced footage, or when using the Fields, Upper First or Fields, Lower First de-interlacing modes. All other de-interlacing modes should be avoided used because they are likely to corrupt the scanlines and prevent the effect from detecting them correctly. Fixing chroma bands can be done using either interlaced or de-interlaced footage with any de-interlacing mode.
If the bands only appear in occasional frames, the Fix Scanline Flicker effect may also be used to remove them. Additionally, If both dark and light bands appear in alternate fields at the same position in a frame then they may not be fixed correctly. In this case, an additional Fix Scanline Flicker effect may be employed to correct the artifacts.
Band Type: The type of bands (dark and/or light) that will be fixed in the luma channel.
Min Length %: The minimum number of pixels (as a percentage of image width) on a scanline that must pass the detection test for a luma band.
Fix Luma: When selected, scanlines that pass the minimum length test above will be fixed.
Max Height: The maximum height of a chroma band, measured in pixels.
Min Length %: The minimum number of pixels (as a percentage of image width) on a scanline that must pass the detection test for a chroma band.
Threshold %: The minimum difference in pixel chroma values for detecting a chroma band.
Fix Chroma: When selected, scanlines that pass the Chroma Bands tests above will be fixed.
The screenshot below shows before and after results when using the Fix Bands effect to fix luma bands, and illustrates the type of artefact this effect has been designed to detect and fix. The bright bands visible in one field of each clip have been corrected.
The screenshot below illustrates the type of chroma band that the effect has been designed to fix. In this case, a band is present in one of the chroma channels and appears as a red stripe in the RGB image.
The Fix Frame effect can be used to fix large gaps or holes in a frame, or even generate completely new frames if they are missing entirely or very damaged. This effect uses information from previous/next frames in the clip to generate missing pixels. Unlike cloning pixels, which just copies pixels from another frame, the Fix Frame effect calculates the correct pixel position using motion estimation to ensure a high quality fix.
The Fix Frame effect can operate in either RGB or YUV colour space according to the colour space parameter set on the clip. The description below is for RGB colour space, but the same principles apply when operating in YUV colour space.
The example below shows a damaged frame on the left, the artifact identified after drawing a polyline around the damaged area in the middle, and the result of fixing using the previous/next image data on the right.
Alternatively, an entire frame that may be missing or too seriously damaged can be replaced.
When fixing over multiple frames, the image data in the previous frame will be interpolated in each frame to match the image data in the next frame. Trackers can be used to assist this interpolation if required.
The Look behind and Look ahead buttons specify which of the previous and next frames will be used to calculate the fix. By default, both are enabled meaning image data will be taken from both the previous and next frames. Either one of these buttons can be de-selected to ensure image data is taken from either the previous or the next frames only.
Motion: The type of motion that will be used to map pixels in the current frame into the next/previous frames.
For example, if the background around a hole is generally translating in X and Y only, change this to Translation. Other options are None for no motion, Horizontal and Vertical to allow translation in those directions only, Trans + Rot to allow only translation and rotation, Affine which will allow small deformations, and Perspective which will allow full perspective planar transformations. This parameter does not affect repair of an entire frame, and is only used when drawing rectangles, lassos or polylines in the image.
Channels: The Red/Green/Blue channels that will be repaired. If a hole or artifact exists in only some of these channels, set this accordingly to improve the quality of the repair.
Deflicker: When selected, brightness variations will be accounted for when fixing.
Grain/Noise Preset: The preset to render grain or noise over the top of the repaired region. In many cases you may wish to de-grain footage before restoration and regrain afterwards. If not this option allows grain/noise to be rendered over the repaired area to prevent boiling when the sequence is played back.
Frame Lookahead: The number of frames before/after the current frame that will be used to repair. In some cases you may wish to increase this to control which frames are being used to calculate the fix. Motion estimation should ensure that any motion within a sequence will be compensated for, but having too much motion between the previous/next frames will reduce the quality of the fix.
Border Size: The size of border or feathering around the boundary that is used to estimate motion into the next/previous frames. The default value is 10%, and increasing this value will mean that more of the image is used to estimate motion. This parameter does not affect repair of an entire frame using the Fix Frame button.
Matte Threshold: The threshold that is used to distinguish between damaged and undamaged pixels after the fix has been performed. The default is 3%, and increasing this value means that fewer pixels inside the boundary will be repaired. This parameter does not affect repair of an entire frame using the Fix Frame button.
Sensitivity: The error threshold for motion analysis. The default is 25%, meaning that pixels which differ by more than this value are considered to be incorrectly tracked, and will not contribute to the overall motion that is used.
Smoothness: The overall smoothness of the motion estimate.
The Frame And Fix lists
Once repairs have been made, the Frame List contains all frames that have been fixed.
Click on a frame number to move to that frame.
All/None: Select all/none of the frames, depending on how many are selected when the button is clicked.
Delete: Delete all repairs in selected frames.
Repaired areas in the current frame are shown in the second list. To select an area, click with the left mouse button in the list. Hold the Shift or Control keys to select multiple areas at the same time.
All/None: Select all/none of the areas in the current frame.
Delete: Delete all selected areas from the frame.
Enable and Disable: Enable or disable the selected repairs.
Marquee: Draw a selectin rectangle to select all areas that intersect a selection in the Cinema window.
Highlight and Names: Highlight and display names are drawn for each repaired area.
Fix Scanline Flicker
The Fix Scanline Flicker effect can be used to correct small variations in the overall brightness and colour of entire scanline in video footage. The effect operates in YUV colour space. The screenshot below shows before and after results of correcting scanline flicker, and illustrates the type of flicker that this effect has been designed to fix.
Channels: The channels of the luma (Y') or chroma (U and V) that contain scanline flicker.
Frame Window: The number of frames on either side of the current frame which will be used to identify and correct scanline flickering. Increasing this value may provide a more accurate result but will also increase the time it takes to fix each frame. The default value is 3, meaning the 3 frames before and the 3 frames after the current frame will be used to correct flicker.
Y', U and V Sensitivity %: The maximum variation in intensity in each channel that is caused by flicker. Increasing these values will correct larger amounts of flicker. The default value for each channel is 10%.
Constant: When selected, the parameters should be held constant over the entire clip or allowed to vary. When parameters vary, their F-Curves are available for editing in the F-Curve editor and Dope Sheet.
The image below shows before and after results when using the Fix Scanline Flicker effect:
Fix Scanline Sync
The Fix Scanline Sync effect can be used to correct scanlines that are incorrectly synchronized with the rest of the image due to timing errors. The effect operates in YUV colour space. The screenshot below shows before and after results of fixing scanline synchronization and illustrates the type of error that this effect has been designed to fix.
Channels: The colour channels of Y', U and V which will be fixed.
Mode: The method used to fix scanlines that are determined to be out of synchronisation. When Shift is selected, scanlines will be shifted horizontally back into place. When Replace is selected, the entire scanline will be replaced by averaging the scanlines above and below.
Interlaced: When enabled, scanline synchronisation will be determined assuming the footage is interlaced. If the footage has been de-interlaced, switching this option off may improve results.
Y', U, V Min Shift and Max Shift: The minimum and maximum shift (measured in pixels) that is present in scanlines. These values should be set according to the defects present in the image. The default values are 1 and 6 pixels respectively.
Iterations: The number of interations that will be used by the solver to synchronise scanlines. When the image contains many scanlines that require fixing, increasing this value will improve the overall results. The default value is 2.
The image below shows before and after results when using the Fix Scanline Sync effect:
The Fix Scratch effect can be used to fix significant horizontal scratches in a clip that might be persistent from frame to frame. Various combinations of bright and dark scratches can be identified automatically and fixed in any of the red, green and blue channels.
Note: The number of CPUs that are used to perform scratch fixing is specified using the Use CPUs For Processing option in the General section of the Preferences window.
These controls can be used to automatically identify and fix significant scratches in the clip.
The Detection order specifies the type of scratches to look for. The options are Bright and/or Dark. Bright and Dark refer to the relative brightness of scratches pixels against the background. If only dark or light scratches are present in the clip, it is recommended to set this parameter correctly to improved the accuracy of the scratch detector. Use one of the Bright then Dark or Dark then Bright options if both types of scratches are present. The order of detection (Bright first or Dark first) matters in cases where for example Bright scratches are embedded into dark scratches.
The View button in the results graph will show an overlay in the Cinema window indicating which pixels are identified as being scratched, given the current detection parameters. This can be used to check that the current parameters are suitable for detecting scratches in the clip. If the overlay is visible whilst parameters are being adjusted, the preview will update automatically. To change the colour of the overlay, right-click on the button and use the colour editor to generate a new colour.
Bright and Dark Scratches
Detect Channels: The R/G/B which channels will be used to detect scratches. Depending on the amount of noise in the footage, it may be advantageous to detect scratches in specific channels, even though the scratches require fixing in all channels.
Fix Channels: The R/G/B channels to fix scratches.
Min Thresh % and Max Thresh %: The contrast range that a scratch must have against the background. Here, contrast refers to the difference in pixel intensity in the range [0..100%]. Decreasing the Min Thresh. parameter will mean that more faint scratches are detected.
Min Width and Max Width: The width of the scratches (the distance between the left and right-hand edge of a scratch). It is important to set this value correctly to ensure that scratches are fixed entirely. If the scratch width is too small, the entire scratch may not be removed from each frame.
Min Height and Max Height: The minimum and maximum height that of a scratch in order to be detected.
Dilate: The number of pixels each scratch will be expanded horizontally by before fixing. This can be used to increase the horizontal size of each scratch slightly and improve the quality of the fix.
The Fix Streaks effect can be used to fix video dropouts that appear as dark or light streaks (sometimes known as "comet tails") in the luma channel. Typically, dropouts can also occur in the chroma channels causing spikes of colour. The Fix Streaks effect works in YUV colour space.
Note that in some situations, the Median Filter effect can also be used to remove these types of artefacts by setting an appropriate vertical filter size and increasing the minimum change required to apply the filter.
The screenshot below shows before and after results when using the Fix Streaks effect, and illustrates the type of artefact this effect has been designed to detect and fix. In this case, dark streaks in the luminance channel and spikes of colour in the chroma channels.
Note that streaks caused by signal dropout affects pixels along a single scanline. Applying any sort of lossy compression to the original footage is likely to disrupt these artefacts and prevent them from being detected correctly.
This effect can operate with interlaced footage, or when using the Fields, Upper First or Fields, Lower First de-interlacing modes. All other de-interlacing modes should be avoided used because they are likely to corrupt the dropouts and prevent the effect from detecting them correctly.
Luma Streak Type: The dark and/or light type of streaks that will be detected.
Luma Streak Level %: The luminance level that dark streaks must drop below before being detected (or equivalently, the luminance level above which light streaks must rise). Increasing this value will mean more feint streaks are detected, but may also increase the number of false positives. The default value is 30%
Luma Streak Threshold %: The minimum change in luminance that will be fixed for each streak. Decreasing this value will mean more pixels are detected, but might also increase the number of false positives. The default value is 5%.
Luma Streak Length: The minimum length of dropouts (measured in pixels) in the luma channel that will be fixed. The default value is 2.
Chroma Threshold %: The minimum change in the chroma channels that will be detected as a dropout. Decreasing this value will mean more chroma pixels are fixed, but may also increase the number of false positives. The default value is 5%.
The image below shows before and after results when using the Fix Streaks effect:
The Fix Tear effect can be used to remove visible tear or splice artefacts in a clip. To fix a tear, the torn part of the film must have been spliced back into approximately the correct position. If this is not the case, the Fix Frame effect may be a better option. The Fix Tear effect will automatically re-align the torn part of the film, removing any artifacts around the tear edge as part of the process.
To fix a tear, you must first identify the frames where the tear is visible and then draw a polyline boundary around the region before the tear and the region after. The region before the tear is the area of the frame that is in the correct position, i.e. lines up with the other frames.
Using the arrow keys to move back and forward in the sequence is often the easy way to see which area of the frame is in the correct position.
The example below shows a torn frame on the left, and the red and green After Tear and Before Tear polylines drawn correctly. In this example, the top part of the frame is already correctly aligned with the previous frame, so the Before-Tear line was drawn at the top, and the After-Tear line at the bottom.
The Fix Tear controls are shown below:
Motion: The type of motion that is needed to re-align the torn part of the frame. For example, if the torn region requires translating in X and Y only, change this to Translation. Other options are None for no motion, Horizontal and Vertical to allow translation in those directions only, Trans + Rot to allow only translation and rotation, Affine which will allow small deformations, and Perspective which will allow full perspective planar transformations.
Channels: The Red/Green/Blue channels that will be repaired.
Deflicker: When selected, brightness variations will be accounted for when fixing.
Grain/Noise Preset: The grain or noise preset to be used to render grain or noise over the top of the repaired region. This can often improve the quality of the fix in situations where the repaired tear region is too smooth.
Frame Lookahead: The direction and frame that the tear will be re-aligned to. By default, this is set to -1 indicating that the tear will be aligned to the previous frame. This can be changed to any positive or negative value, where positive values indicate that the next frames will be used, rather than the previous frames.
Border Size: The size of the region around the tear that will be used to estimate motion when removing the tear artefacts. The default value is 10% of the overall image size. Increasing this value will mean that more of the image is used to obtain a motion estimate.
Matte Threshold: The threshold used to decide if a pixel is corrupted by the tear or not. The default value is 3%, indicating that a repaired pixel must differ from the original by at least 3% to be included in the tear fix.
Sensitivity: The error threshold for motion analysis. The default is 25%, meaning that pixels which differ by more than this value are considered to be incorrectly tracked, and will not contribute to the motion estimate.
Draw Before Tear and Draw After Tear: Draw the polyline around the areas before the tear and after the tear. To place a vertex of the polyline, click with the left mouse button in the Cinema window. The Before-Tear polyline is drawn in green and the After-Tear polyline in red. To close the polyline, click on the first vertex again. Once you have drawn one polyline, the buttons will switch state, allowing to to immediately draw the second polyline. Make sure that the tear itself (including any artefacts surrounding it) is contained entirely between the two polylines. If there is no part of the image before or after the tear, there is no need to draw a polyline in the frame. A polyline can be edited after it's drawn by clicking and dragging on a vertex with the left mouse button.
Clear Boundary: Remove all the polyline boundaries in the current frame.
Start: Start the tear fixing process, repairing the tears in all frames that have a polyline boundary. For the example above, this results in the torn part of the frame being moved back into the correct position:
The Tear list
Once a tear has been fixed, the Frame list contains all frames that have been corrected.
Clicking on a frame number will move you to that frame. Click All/None to select all/none of the frames, and click Delete to delete all repairs in selected frames.
Repaired areas in the current frame are shown in the right-hand list. To select an area, click with the left mouse button in the list. Hold the Shift or Control keys to select multiple areas at the same time. Click the All/None button to select all/none of the areas, and click Delete to delete all selected areas. The Enable and Disable buttons can be used to enable or disable the selected repairs. To select repaired regions from the image, click the Marquee button and draw a selection rectangle in the Cinema window using the left mouse button. The Highlight and Names button are used to ensure highlights and names are drawn for each repaired area.
The Lens Un-Distort effect will correct for any camera lens distortion in your clip.
Mode: The distortion model that will be corrected. Choices for the distortion model are Low for low-order radial distortion only, Low + High for low and high-order distortion, and Low + High + Offset to include an offset to the centre of lens distortion (measured in pixels). Each parameter is controlled by a slider and edit box.
Update: When selected, provide interactive updates in the Cinema as each parameter is adjusted.
Display Grid: Draws an overlay of a rectilinear grid in the Cinema to help visualise how much lens distortion is being removed.
Constant: When selected, the distortion parameters are constant over the entire clip. With this set, each adjustment to the distortion control sliders will change the distortion parameters for the entire clip. If distortion is not constant, it can be adjusted independently for each frame. The distortion parameters can also be edited using the F-Curves editor and Dope Sheet.
To automatically estimate the lens distortion parameters, click the Find Edges button. This will find significant edges in the current frame and display them in the Cinema. Edges can be selected by clicking on them with the left button, or by dragging a selection marquee with the left button after clicking the Marquee button. To select all/none of the edges at once, click the All/None button. Selected edges are drawn in yellow. After selecting one or more edges, clicking the Solve button will estimate the distortion parameters to straighten those edges. Selected edges can be deleted by clicking the Delete Edges button.
To manually draw a distortion edge, click the Draw Edge button and use the left mouse button to draw a line in the Cinema. Click with the left button and drag the mouse, releasing the left button to place the end-point of the line. Once a line is created, it can be moved by clicking and dragging the end-points. Additional vertices can be inserted by clicking and dragging with the left mouse button. In order to estimate distortion parameters from a manually drawn line, it must contain at least 3 vertices and should be positioned to follow a straight edge in the clip.
Manual Dirt / Dust Fix
The Manual Dirt/Dust Fix effect provides tools that can be used to manually identify dirt areas for cleaning, using either temporal or spatial filling. Rectangle, lasso and paint tools are available for manually specifying areas of the image that require cleaning.
These tools are intended to detect and fix artefacts that appear in a single frame only, such as spots of dirt or dust, and can be cleaned using either a temporal or spatial filling algorithm.
The Motion Analysis box contains parameters that affect the accuracy and speed of the motion analysis required for dirt detection or temporal cleanup.
Accuracy can be used to set the accuracy of motion estimation when performing dirt detection and temporal cleanup. Choices are Low, Normal and High. Setting this to low will mean that motion estimation is performed faster than normal, but at the cost of decreased accuracy.
Lookahead: The number of frames to search behind and in front of the current frame when detecting and fixing dirt pixels. By default this is set to 1, which means that pixels from the previous frame and the next frame will be used. With a lookahead value of 1, dirt can only be detected if it persists for a single frame.
< and >: Which of the previous and next frames will be used to fix dirt marks temporally. By default, both are enabled meaning image data will be taken from both the previous and next frames. Either one of these buttons can be disabled to ensure image data is taken from either the previous or the next frame only.
Smoothness: The overall smoothness of the motion field to be controlled. Increasing this value will mean that smoother motion fields are generated, possibly at the expense of decreased accuracy around edges of moving objects.
Block Size: The block size (measured in pixels) for motion analysis. A smaller block size can provide more accurate motion data, but at the expense of increased computation time.
The Detection and Fix parameter options are very similar to those in the Auto Dirt/Dust Fix effect.
Contrast: The relative sensitivity of the dirt detector inside the area drawn by the user. This means that the Manual Dirt/Dust Fix effect can be used to clean areas that cannot be detected or fixed automatically using a single set of parameters for the entire frame.
Channels: Which of the R/G/B channels will be used to detect dirt. In some cases, you may get better results by detecting dirt in fewer channels but fixing in all channels. For example, if the blue channel contains a significant amount of noise compared to red and green, it will be easier to identify dirt areas if the blue channel is ignored during the detection process
Use Defect Map: The data source used to find dirt pixels. If the clip contains a pre-generated defect map the button becomes active. Press the button to use the defect map to detect dirt. With the button greyed out or not pressed, image pixels will be used to identify dirt (see the Defect Map section in the documentation for the Clip Panel and Auto Dirt/Dust Fix effect).
Reset: Reset the Detection parameters to their default values.
Spatial and Temporal: The method used to fix dirt pixels after they have been identified. Temporal will use pixels from nearby frames, and Spatial fixes will use nearby pixels from the same frame. Selecting both will use a threshold (Temp. Thresh) to switch automatically between spatial and temporal fixes based on the size of the fix. Pieces of dirt that are larger than this threshold will be fixed temporally, while dirt smaller than the threshold will be fixed spatially.
Blend Radius: The size detected dirt areas to be expanded and blended with the background image after cleaning, reducing visible artefacts around the edges of each area. To change the blend radius of existing areas, adjust this value and then click the Re-Fix Areas button described below.
Channels: Which channels will be fixed.
Deflicker: When enabled, will attempt to compensate for large scale illumination differences between frames when fixing areas.
Grain/Noise Preset: The preset that is used to render grain or noise over the area that the object is removed from. This can be used to ensure that areas cleaned using a spatial filling algorithm do not appear too smooth compared to the rest of the image.
Reset: Reset the Fix parameters to their default values.
Dirt Area Controls
To specify an area for cleanup, click one of the Rectangle, Lasso or Paint buttons, and draw in the cinema using the left mouse button. Click and drag with the button held to define a rectangular area, a free-form lasso, or a paint stroke. To change the size of the paint brush, hold the Shift key down whilst dragging the mouse with the left button.
A keyboard shortcut is available (V by default) that will mark a piece of dirt at the current cursor position using the most recent rectangle, lasso or paint stroke. Another keyboard shortcut (Shift+V by default) can be used to remove the piece of dirt that is nearest the cursor.
The lists of cleaned frames and dirt areas behaves in the same way as those in the Auto Dirt/Dust Fix effect, and that documentation can be referred for the function of the various buttons and lists.
If a cleaned area is selected from the list, the Toggle Fix Mode button can be used to change between a spatial and temporal fix for that area. For temporal fixes, the Move Temp. Fix button can then be used to change the horizontal or vertical offset that has been used to generate the fix. In certain situations, motion analysis cannot be performed accurately, and a temporal fix may appear offset from the surrounding image by a few pixels. If the Move Temp. Fix button is pressed, the cursor keys can be used to scroll around the image and correct the motion offset that was used to generate the cleaned pixels.
Grow Area: Dilate an area by one pixel in each direction. This can be used to correct situations where an area of dirt has only been partially detected.
The Manual Dirt/Dust effect has the ability to re-fix dirt areas in one or all frames automatically. This can be useful when the underlying clip has been changed (for example, using colour correction) because it allows all existing dirt areas to be updated without the need to manually mark them again. Clicking the Re-Fix Areas button will re-fix all selected areas in the current frame. Clicking the Re-Fix Frame button will re-fix all areas in the current frame, and clicking the Re-Fix All button will re-fix areas in all frames.
The Median Filter effect applies a median filter to each channel of the clip.
The Median Filter effect can operate in either RGB or YUV colour according to the colour space parameter set on the clip. The description below is for RGB colour space, but the same principles apply when operating in YUV colour space.
The horizontal and vertical filter sizes (in units of pixels) can be changed using the Red Size, Green Size and Blue Size controls. The Gang XY tick-box controls whether these sizes are adjusted together or individually. Reducing the size to 0 in each direction will disable the filter in the corresponding channel.
Gang RGB: When selected, all red/green/blue channels will be adjusted at the same time.
Constant: When selected, the filter values are always constant throughout the clip. The median filter parameters are available for editing in the F-Curve editor and Dope Sheet.
Temporal: When selected, a temporal filtering algorithm will be used, including pixels from adjacent frames in the filter. The temporal median filter can give superior results in many situations, but will increase the processing time for each frame.
R Min %, G Min % and B Min %: The minimum change to each pixel that will be applied (default is 0%). After applying the median filter, if the difference in pixel intensity is less that this value, the filter will be ignored at that pixel. Increasing these values will mean that small details in the image are not removed by the filter.
R Similarity %, G Similarity % and B Similarity %: The range of pixel values that will be included in the median estimate (default is 100%). When applying the median filter at a pixel, only surrounding pixels that are similar to the central pixel will be included in the median calculation. Decreasing these values will mean that significant elements in the image are not affected by the median filter.
The percentile used to find each median value is controlled using the R %, G % and B % sliders and edit boxes. By default, this is 50% meaning that the median is taken to be the mid-point of all pixels included in the filter.
The filtered image will be blended with the original according to the Blend parameters for each channel.
The Noise Removal effect provides a fast alternative approach to reducing or removing high frequency noise artefacts from a clip whilst preserving important image information such as gradients and edges.
Noise Removal %: The amount of noise to remove. Raising these values up to a maximum of 100% will mean that increasingly significant pixel colour changes are classified as noise.The default value for each channel is 2%.
After setting Noise Removal parameters that are suitable for removing all the noise from a clip, the Detail Protection parameters can be adjusted to ensure that significant image features such as edges are preserved. The default Detail Protection value is 90, but it can be increased above 100 where necessary. If the Detail Protection values are set to zero, the noise removal will act more like a low-pass filter.
Gang: When selected, parameter adjustments are applied to all channels at the same time.
Constant: When selected, parameters remain constant throughout the clip.
Colour Space: The colour space in which noise removal is performed. The default is YUV, but this can also be changed to RGB, YCbCr or YIQ when required.
The image above shows before and after examples of the Noise Removal effect in action.
The Paint effect allows paint strokes to be placed manually throughout a clip. Paint strokes can be applied to each frame, and they can also be animated and tracked through the clip. To paint a stroke, click an Action button and use the left mouse button to paint in the Cinema. Paint strokes that are present in the current frame are displayed in the Paint Stroke Table, on the right of the panel.
There are various types of paint actions that can be used to generate paint strokes. Brush actions are controlled by clicking on one of the Action buttons in the paint action palette.
Note that brush parameters such as size, opacity etc.. are remembered for each type of brush action. Clicking on a brush action button will update the brush parameters accordingly.
Paint will paint the brush colour into the image.
Tint will tint the image to give it the same hue and saturation as the brush colour.
Blur will apply a Gaussian blur to the image over the pixels the paint stroke is applied to.
Sharpen will apply a sharpen filter to the image.
Smooth will smooth noise in the image whilst trying to maintain edges.
Grain will apply de-grain and/or re-grain operations using a grain or noise preset that can be chosen using the menu underneath the paint stroke list. This button is not available until a grain or noise preset is chosen.
Clone will clone from another part of the image. To use the clone tool, press and hold the Control key and click with the left mouse button to specify the position to clone from. Drag the mouse whilst the left button is held, and release the button at the position to clone to, and release the Control key.
The brush is then applied in the usual way, by clicking and drawing with the left mouse button. To display the clone source image as a semi-transparent overlay (onion skin effect), press and hold the Option key. The clone source can be transformed using the transform widget that appears on screen. To translate the overlay, click and drag on the circle with the left mouse button. Dragging the inner and outer points on the circle allow the clone source to be rotated and scaled. Dragging the corners and edge of the clone source boundary allows perspective and non-uniform scaling to be applied. Releasing the Option Key will hide the overlay from view.
To clone from one frame to another, use the Control key to specify a clone position, as described above, then change frame, and use the Option key to transform the clone source into position. The frame number or timecode of the source frame can also be typed into the Source frame text box. To adjust the transparency of the overlay, press the - (minus) and + (plus) keys whilst holding the Control key. When holding down the Option key to display the clone source frame, the cursor keys can be used to adjust the translation of the source by one pixel in any direction. If the Shift key is held as well, the translation will be updated by 5 pixels in any direction.
Colorize will clone colour (hue and saturation) from another part of the image. The Colorize brush operates in the same way as the Clone brush, described above.
Equalize will clone luminance from another part of the image. The Equalize brush operates in the same way as the Clone brush, described above.
Repair will attempt to fill pixels using other pixel colour from the same frame. This is useful for removing blemishes or unwanted artefacts from the image. By default, the repair brush will use nearby pixel colours to fill in the painted area. To change which pixels are used, specify a source area by holding the Control key and drawing a rectangle in the Cinema with the left mouse button.
Clean will fix pixels using pixel colours from the previous/next frames after motion compensation. This is useful for removing blemishes or unwanted artefacts from the image, and can give superior results to the Repair brush in many situations. If motion data is not available for the frame, it will be calculated in the necessary area.
Dodge will increase the apparent exposure of painted pixels.
Burn will decrease the apparent exposure of painted pixels.
Erode will remove darker pixels by using nearby lighter pixels. This can be useful for painting out small dark scratches.
Dilate will remove lighter pixels by using nearby darker pixels. This can be useful for painting out small light scratches.
Multi-Frame Paint Strokes
By default, paint strokes are active in the frame into which they are first drawn. To draw paint strokes that are applied in all frames (for example, to fix a persistent piece of dirt using the Repair brush), click the Multi-Frame button in the brush controls before drawing a new paint stroke.
Paint strokes that exist across multiple frames are available in the F-Curve and Dope-Sheet editors for further editing.
Multi-frame paint strokes can either be tracked directly inside the paint editor (see below) or attached to existing trackers that are generated from the Tracker panel.
The current brush size, shape and opacity is shown as an icon in the Brush palette. The brush palette can be opened by clicking the B button in the paint action palette.
The brush Mode can be changed to:
Brush to paint with the current brush shape
Rectangle to paint a rectangular shape
Ellipse to paint an elliptical shape
Line to paint a straight line
Lasso to lasso pixels for painting.
The Shape buttons control the shape of the brush. They are available in Brush Mode and Line Mode. Circular, rectangular and line brush shapes are provided.
Channels: The red, green and blue that channels will be painted.
Multi-Frame: When selected, newly painted strokes exist throughout the entire clip, rather than just on the current frame.
Falloff Profile: Open an editor window which allows the radial density profile of the brush to be set by adjusting a graph. The X (horizontal) axis represents radial distance from the centre of the brush with the centre on the left and the outer edge on the right. The Y (vertical) axis represents density with opaque at the top and transparent at the bottom. The default falloff profile results in a brush whose density decreases uniformly between the centre and the outer edge
To insert a new control vertex, click on the curve with the left button, or click with the right button to remove a vertex. To make a linear control vertex, drag the Bezier handles together, and hold the Shift key when moving a vertex to expose the Bezier handles again.
Size: The size of the brush. Brush size can also be set interactively in the Cinema by holding the Shift key and dragging with the left mouse button held.
Opacity: The brush opacity. Brush opacity can also be set interactively in the Cinema by holding the Shift key and dragging with the right mouse button held.
Pressure: The effect of tablet pen pressure to the brush stroke. This can be set to Opacity to change brush opacity, Size to change brush size or Opacity + Size to change both.
Pressure Profile: Open an editor window to specify how pressure changes from the pen tablet affect brush size and/or opacity.
By default, the relationship is linear, meaning that twice as much pressure will double the size or opacity of the brush. Click and drag the control vertices with the left mouse button to change the profile. To insert a new control vertex, click on the curve with the left button, or click with the right button to remove a vertex. To make a linear control vertex, drag the Bezier handles together, and hold the Shift key when moving a vertex to expose the Bezier handles again.
Store: When selected, each paint stroke that is drawn will be stored before being merged together and listed in the Stroke table as a single paint stroke. The button label will change to Apply, and when pressed again all stored strokes will be merged together and placed in the Stroke List. The Store button can also be used to extend existing paint strokes (see below).
Defect Map: When selected, paint strokes should only cover pixels identified by the clip's defect map (if available). The Show Defects button can be used to display the contents of the defect map. Clicking with the right mouse button will display a colour chooser to adjust the colour of the defect map overlay.
Source frame: The source frame for a clone brush. Source frame and the Lock option are available when using a clone brush. When the Lock option is un-checked, the clone source frame will be changed when the target frame (i.e. The frame that is being painted) is changed. The offset between the target frame and the source frame is defined according to the first clone brush stroke. For example. If the clone source is grabbed at frame 1, and then a brush stroke is painted at frame 10, painting again at frame 11 will mean the clone source frame is changed to frame 2.
Deflicker: When enabled, the clone and clean brushes will attempt to compensate for local illumination changes when painted. This can be useful when cloning or cleaning from frames that are significantly darker or lighter than the current frame. This option is available for the clone and clean brushes.
Presets for brush properties may be saved and recalled via the Brush presets list, highlighted in red in the screen shot below.
Add: Store the current brush properties. All properties of the current brush within the Brush and Colour tabs will be stored.
Recall: Use the brush properties stored in that preset. All properties of the preset within the Brush and Colour palettes will be set to the values stored in the preset. Double-clicking on a preset allows the preset to be renamed.
Del: Remove the selected preset from the list.
The colour controls can be used to change the current brush colour. The colour controls palette can be opened by clicking the C button in the paint action palette.
The colour of the brush is only important for Paint and Tint brush actions. Click in the Colour Wheel to pick a colour, or drag the marker in the horizontal Luminance Bar to change the luminance of colours in the colour wheel. The current brush colour is shown in a square at the bottom-left of the colour controls.
Reset: Reset the colour wheel to its default state (white brush colour).
Pick: Allow a colour to be picked from either the Cinema window or from the Colour Mixer.
The Colour Mixer can be used to mix colour together. Paint into the mixer with different brush colours using the left mouse button. To mix colours together, press the Mix button and then use the left mouse button again to mix the colours together. Clicking the Clear button will clear the mixer and change the background colour to the current brush colour.
The Colour Swatches below the mixer are used to store preset colours. To set the brush colour to one of these presets, click on a swatch with the left mouse button. To store the current brush colour in a preset, click on a swatch with the right mouse button. To change a preset colour using a colour chooser, double-click on a swatch with the left mouse button.
Paint Stroke table
Each paint stroke is stored in the table when it is created.
The table contains all strokes that are present in the current frame, along with multi-frame paint strokes. The strokes are listed in the order in which they were drawn in the image, with newer strokes at the top of the table.
To rename a paint stroke, click on its name with the left mouse button. The Action column shows the brush action that was used to create the stroke.
The In and Out columns indicate the in and out frames for the stroke. These can be changed by editing the values after clicking in the column with the left mouse button of by using the In/Out Frame buttons below the stroke table
When trackers are available, multi-frame paint strokes can be attached by right-clicking in the Tracker column and selecting a tracker name from the popup menu.
The remaining columns in the table are used to control how the paint stroke is applied to the image. Each entry can be switched on/off by clicking with the left mouse buttons. Active entries are indicated by a small cross (X) in the table. The column names correspond to:
Co: Allow the contrast of the paint stroke to change. If this column is switched on when the stroke is tracked, the contrast change will be estimated automatically from the image data.
Br: Allow the brightness of the stroke to change. If this column is switched on when the stroke is tracked, the brightness change will be estimated automatically from the image data.
Mb: Render motion blur as the stroke is animated. The Motion-Blur F-Curve represents the camera shutter-angle, and default to 180 degrees. When rendering motion-blur, the distance travelled by each pixel will be calculated using the paint stroke transformation in each frame.
Dg: Remove grain from the paint stroke (using the current grain or noise preset).
Rg: Render additional grain over the paint stroke (using the current grain or noise preset).
Dy: When enabled, the paint stroke updates itself dynamically as it is animated. This can be used to ensure, for example, that a Repair stroke repairs separately in each frame instead of repairing once and copying that data into other frames. Not all brush actions can be set to dynamic updates. If an action cannot be applied dynamically, a small - symbol will appear in this column.
Note that the Dg and Rg columns will not be available until a grain or noise preset has been chosen using the menu underneath the stroke table.
To select a stroke, click on the stroke in a row with the left mouse button. All selected strokes have a yellow border drawn around their edge. You can change the column entries described above for multiple strokes at the same time by selecting several rows in the table (hold the Shift or Control keys to do this).
All/None: Selects all/none strokes in the list at the same time.
Enable: Enable all selected strokes. Enabled strokes will be drawn into the image.
Disable: Disable all selected strokes. A disabled stroke does not draw itself into the image.
Marquee: Allows strokes to be selected in the Cinema, by drawing a rectangle with the left mouse button. Holding the Control key whilst drawing the marquee will mean a free-hand lasso can be used instead of a rectangular region to provide more accurate selections.
Merge: Merge all selected strokes into a single paint stroke.
Delete: Delete all selected strokes from the effect.
Make Multi: Change the selected single-frame paint strokes so they exists in all frames.
Make Single: Change the selected multi-frame paint strokes so they only exist in the current frame.
The in/out frames of a stroke can also be changed by either editing the In/Out columns in the stroke table or by pressing the In/Out Frame buttons underneath the table.
Highlight: Toggle display of the yellow highlight around each selected stroke that is drawn in the Cinema window.
Hide and Show: Hide or show multi-frame paint strokes in the current frame. A hidden paint stroke will not be rendered into the frame.
Editing paint strokes
Once a stroke is selected, press and hold the J key in the Cinema to display the transform tool. The stroke shape can be manipulated by dragging the corners of the transform tool with the left mouse button. Releasing J will paste the stroke back into the image (Note: this keyboard shortcut can be changed in the preferences window if desired).
Extending Paint Strokes
A paint strokes can be extended to cover more pixels by first selecting the paint stroke by clicking in the paint stroke table, and then pressing the Store button. Whilst painting in the Cinema, the stroke will be extended to include newly painted pixels. Clicking Apply will then update the paint stroke accordingly.
Tracking Paint Strokes
To track a paint stroke, the stroke must first be active in multiple frames. If motion data is available, it will be used to track the stroke. If no motion is available, it will be calculated in the necessary areas as needed by the paint stroke tracker.
To start tracking a paint stroke, make sure it is selected in the table, and click either the >> or << buttons to track the stroke forwards or backwards through the clip. To stop tracking at any time, press the Escape key or click the button again. To track one frame at a time, use the < or > buttons. To change the motion model that is used to track a stroke, click on the menu box underneath the stroke table to choose from Translation, Affine or Perspective. Translation will track only x and y position. Affine will track with 6 degrees of freedom, allowing for small changes in orientation and scale. Perspective will track with a full planar-perspective model of motion, with 8 degrees of freedom.
As well as tracking position and orientation, changes in brightness and/or contrast and also be tracked. To do this, make sure the Co and/or Br columns are enabled in the table before tracking.
Pan And Scan
The Pan And Scan effect can be used to translate, rotate and scale an individual clip. Note that the Pan And Scan effect does not resize the clip - for that functionality see the remastering facilities in the Remaster node, or this can be applied globally in a Standards conversion node.
The clip can be transformed using the various controls in the node editor, or by mouse manipulation of the clip in the Cinema.
1 Drag the centre box within the circle to move XY
2 Drag the boxes at the left or right of the circle to move X
3 Drag the boxes at the top or bottom of the circle to move Y
4 Drag the corner boxes for uniform scale
5 Drag the side boxes to adjust aspect ratio
When Aspect or Rotation have been changed, the Snap to options in the right hand column determine how the frame is filled.
The Method drop-down offers a choice of scaling algorthms.
The Re-Grain effect uses grain presets to add grain to a clip.
Mode: Re-grain each channel separately (RGB) in each of the red, green and blue channels, or all channels of the image are re-grained together using a single grain pattern (Grey).
Size: The size (in pixels) of the grains that are rendered into the clip.
Intensity: The grain intensity in each channel, in the range [0..100].
G (Gang): When selected,apply edits to all R/G/B channel values at the same time when pressed.
Seed: The random number seed that is used to generate the grain pattern.
Low, Mid and High: Increase or decrease the amount of re-grain for low, mid and high intensity values in the image. The regions for low, mid and high can be adjusted using the pivots in the profile graph.
Level: The amount of grain added to the clip. The default value is 1.0 and corresponds to adding the most grain (as defined by the parameters) from the clip.
The Preset palette allows a pre-defined grain preset to be used. Selecting a preset from this menu will change the re-grain parameters accordingly.
The Re-Noise is used to add noise to a clip.
Intensity: The noise intensity in each channel, in the range [0..100].
Mode: Re-noise each channel noised separately (RGB) in each of the red, green and blue channels, or whether all channels of the image are re-noised together using a single noise pattern (Grey).
Gang: When selected, apply edits to all R/G/B channel values at the same time.
Level: The amount of noise that is added to the clip. Menu options are High, Medium and Low. The default value is High and corresponds to removing all noise (as defined by the parameters) from the clip.
Seed: The random number seed that is used to generate the noise pattern.
The Remove Rig effect can be used to automatically remove a moving object from a clip, and fill the pixels using image data from other frames of the clip. As well as removing objects like microphone booms or unwanted objects on the set, this effect can also be used to remove defects like cue-marks from the clip, by replacing them with pixels from frames that do not contain such a mark.
Objects can be also be removed using an additional clean plate clip.
To remove an object, first create a mask around the object using the Mask tools. The mask must cover the object entirely and only be present in the frames where the object is visible. The Mode menu specifies the type of motion transformation that will be used to remove the object (see below).
Removing an object without a clean plate clip
Rig removal without a clean plate clip works best when the motion of the background that the object is moving against can be modelled using No Motion, or a Translation, Affine, or Planar-Perspective transformation. Translation assumes the entire background is moving in only the X and Y directions. Affine allows the background to undergo small rotations or scale, and Planar-Perspective assumes that background is a flat plane undergoing a perspective transformation.
The Mode menu specifies the type of transformation that will be used. This should be set according to the motion of the background compared to the object. For example, if the background is simply translating in X and Y, set this to Translation. If the background is moving in a more complex way, set Mode to Affine or Planar-Perspective.
Removing an object with a clean plate clip
To remove an object using pixels from a clean-plate clip, drag the clip out of the Media Bins and drop it into the Clean Plate Clip window in the Remove Rig panel.
The clean-plate clip must be the same resolution as the clip you are currently working on. Changing frames in the Cinema will update the thumbnail displayed in the window.
To hold the clean plate clip at the current frame, click the Lock button. The Remove button can be used to remove the clean plate clip from the effect. To display the clean plate clip in the Cinema, click the Show button.
You can adjust the transparency of the clean plate clip when it is displayed in the Cinema using the Blend slider. Clicking on the Colour Swatch underneath the slider will display a colour chooser that allows you to change the colour of the clip. Note that changing the colour and transparency of the clip only affects the way it is displayed in the Cinema - it will not change the results of the rig removal process.
The thumbnail in the Clean Plate Clip window illustrates the frame that will be used to remove pixels contained inside the mask in the current frame. The frame numbers used can be adjusted from within the F-Curve editor. A parameter named Clean Plate Timewarp Value can be used to control the mapping between frames in the working clip and frames in the clean-plate clip.
The clean-plate clip should be shot from a position that is as near as possible to the original frame. The nearer the camera position, the easier it will be to align the clean-plate clip with the original and remove the object. The Mode menu is used to define the motion model that will be used to warp the clean-plate clip into the correct position. As well as the Translation, Affine and Planar-Perspective options described above, Free Motion is also available when using a clean plate clip. When using this mode, PFClean will attempt to estimate the correct transformation without assuming it fits one of the other models. This can be used when the other motion models are not suitable, although will increase the complexity of the rig removal process.
Trackers may be used to provide hints to the motion estimation that is required during the rig removal process.
Trackers are used to indicate how particular pixels in a clip are moving and they can be tracked in the current working clip or in the clean plate clip (if one is being used).
Trackers should be created in the Tracker panel, and each tracker must also be tracked in the clean plate clip, if one is being used. To do this, click the Show button to display the clean-plate clip in the Cinema and then switch to the Tracker panel. Trackers that have not been tracked in the clean-plate clip will be greyed-out in the tracker list.
When using trackers with a clean-plate clip, clicking the Warp button underneath the clean plate thumbnail will adjust the position of the clean plate clip in the Cinema according to the position of the trackers. This can be used to check the alignment between the clean plate clip and the working clip, to ensure that the trackers are positioned accurately. When using the Translation motion model, this can be done after creating a single tracker. For Affine and Planar-Perspective models, at least 4 trackers are required. The clean plate clip cannot be warped when the Free Motion model is being used.
Rig Mask: The mask that is used to cover the object being removed.
Border Size: The area around the object mask that will be used to estimate background motion. By default, this is set to 5% of the image size. Increase or decrease this value to consider more or less of the background pixels when estimating motion in each frame.
Matte Threshold: The threshold value that is used to decide which pixels belong to the object and which do not. When this value is zero, every pixel inside the object mask will be replaced with one from another frame. The default value of 3 means that a pixel colour must by differ by at least 3% compared to the background before it is classified as being part of the object. This can be used to avoid having to draw a very tightly-fitting mask around the object. Soft-edges masks can also be used to assist with blending the replacement pixels with the original image.
Deflicker: Whether brightness adjustments will be estimated whilst cleaning pixels. This can be useful if the clip contains significant exposure or brightness fluctuations.
Grain/Noise Preset: The preset that is used to render grain or noise over the area that the object is removed from. This can be used to ensure that pixels pulled from other frames to remove the object do not appear to be static when the clip is played.
Clear Frame: Clear all removed objects in the current frame.
Clear All: Clear all frames in the clip.
Start: Start the rig removal process.
Show Outline: When enabled, highlight the pixels that have been replaced in the Cinema window after an object has been removed.
The Sharpen effect applies an unsharp mask to the clip.
Red Size, Green Size, Blue Size: The size (in pixels) of the sharpen filter used to calculate the mask.
Amount: The amount of sharpening that is applied in each channel.
Gang RGB and Gang XY: When selected, all RGB channels or X/Y values are adjusted at the same time.
Constant: When selected, the sharpen parameters remain constant throughout the clip. When this box is not ticked, parameters can be edited for individual frames and are available for editing in the F-Curve editor and Dope Sheet.
The Stabilize effect is used to remove high frequency fluctuations in frame position, or to lock-off all frame motion throughout the clip.
After stabilisation has been performed, the motion curves are available for editing in the F-Curve editor and Dope Sheet.
As well as stabilising over a border region, trackers can be used to assist stabilisation, or specific areas can also be drawn in the Cinema. When one or more areas are placed, the stabilising transform will be estimated using only the pixels inside the areas.
Trackers must be created and tracked using the Tracking Panel and activated in the Stabilize effect. For Translation, Horizontal and Vertical modes, a single tracker is required, although if more are present then their average motion will be used. For Affine and Planar Perspective modes, at least 4 trackers are required. When there are not enough trackers available, modes will appear greyed-out in the Mode menu.
Mode: The type of motion that will be removed. For most situations, Translation, Translate and Rotate, Horizontal or Vertical mode can be used to remove horizontal and/or vertical translation. For small-scale rotation and scale changes, the Affine model can be used (6 degrees of freedom), and Perspective model can be used for full planar-perspective motion (8 degrees of freedom).
Channels: The red, green and/or blue image channels that are stabilised.
Frame Window: The number of frames on either side of the current frame used to smooth out motion fluctuations. The default value is 5, which corresponds to averaging using 5 preceding frames and 5 following frames.
Sensitivity: The sensitivity to differences in pixel intensity that are assumed to be caused by motion (the default value is 15%). Pixel intensity differences above this value are assumed to be caused by incorrect motion estimates and are therefore ignored during the stabilize calculations.
Border: The size of the border around the clip frame that is used to estimate the stabilise transformation. Increasing this value will include more of the frame (the current border area is drawn in purple in the Cinema window). A mask can also be used to specify the area to stabilise. If the alpha channel of the clip contains zero values, then the pixels that have non-zero alpha will be used to stabilise the clip.
Target Frame: The frame to which all others will be stabilised. This frame will not undergo any transformation after stabilisation. The Target Frame button can be used to set the target to the current frame, and the Clear button used to remove the target frame. When no target frame is set, the overall transformation will be estimated so as to minimise the amount of image data that is lost off-screen. By default, the target frame is set to the middle of the clip.
Lock Motion: When selected, remove as much motion as possible from the clip, rather than just smoothing out motion fluctuations within the frame window.
Sub Pixel: When selected, ensure that motion estimation is performed at sub-pixel accuracy. Note that this means the final image may be re-sampled when rendering. With the sub pixel option disabled, the image will only be stabilised to the nearest pixel, meaning that no re-sampling will occur.
Start: Start the stabilisation process.
Reset Motion: Reset the solved motion to zero, allowing it to be resolved after adjusting any parameters as required.
Add Area: Draw an area by clicking and dragging with the left mouse button to in the Cinema.
Marquee: Allow a selection rectangle to be drawn in the Cinema to select areas.
All/None: Select all of the areas or deselect all, depending on how many areas have been selected when this button was clicked.
Delete Areas: Delete selected areas.
Note: The number of CPUs that are used to perform stabilisation is specified using the Use CPUs For Processing option in the General section of the Preferences window.
Video Grade emulates a Telecine-style colour corrector, with Lift, Gamma and Gain controls for the Shadows, Midtones and Highlights
The Wheels control the overall level, and the Trackballs control the colour balance
The Pivots, which can be adjusted in the Response Curve panel, control the range and limits of the Lift and Gain values.
Here, the Gain Pivot, labelled 'G', has been raised above its default location at the bottom of the Response Curve. The Gain adjustment (increased) rotates around it, and has no effect below it.
Here, the Lift Pivot, labelled 'L', has been raised below its default location at the top of the Response Curve. The Lift adjustment (decreased) rotates around it, and has no effect above it.
When the Lift and Gain pivots are offset, they define the upper and lower limits of the Gamma adjustment (increased) as shown here.
When Video Grading is changed to ASC Compatible mode, the Pivots are disabled, as these are not compatible with the ASC CDL specification.
Selecting ASC Compliant allows grading in the industry-standard ASC-CDL colour values, and import / export of .CCC files, enabling look exchange with third-party solutions.
The ASC CDL (American Society of Cinematographers' Colour Decision List) parameters are described below:
Slope Equivalent to Video 'Gain' affects image Highlights Offset Equivalent to Film 'Density' affects image Overall Power Equivalent to Video 'Gamma' affects image Midtones
Looks can be imported to the Video Grade effect by pressing Import and using the browser to select a .CCC file. The current look can also be exported by clicking the Export button and using the browser to select a folder in which to store a .CCC file