Align Cameras
The Align Cameras node has two inputs and two outputs, and can be used to align the camera and trackers from the second input to the same coordinate system as the first input. This can be useful in situations such as:
If two or more camera solver nodes have been used to solve for different parts of one shot, the resulting cameras need to be brought into alignment to create a complete camera path. This can be done automatically if there are two or more frames that have been solved for both cameras, or if three or more solved trackers exist that are common to both inputs.
If two or more cameras are viewing the scene, and each has been solved with a different camera solver, the 3D tracker positions generated by each solver need to be brought into alignment. This can be done automatically if three or more solved trackers exist that are common to both inputs.
Note: for a tracker to be common to both inputs, it must have been tracked into both clips from within the same tracker node. Trackers that have been generated using different tracker nodes will not be common to both inputs, even if they have the same name.
Usage
The Auto Align node can automatically estimate a transformation to align the coordinate system of the second input with the first. The transformation can be adjusted manually using the transform mode and origin menu options and refined to better match the 3D tracker positions and overlapping camera frames if appropriate.
The automatic alignment algorithm will attempt to align camera frames that are common to both cameras in the situation where both cameras are associated with the same input clip, as well as trackers that are common to both inputs. Trackers that are common to both inputs are listed in the tracker list and shown in viewer windows with a red connecting line. Trackers can be activated to include them in the automatic alignment calculation, or deactivated to exclude them.
Controls
Current clip: The clip currently being viewed in the main window. Note that if both camera inputs come from the same clip, this option is not available.
Transform
This displays the current value of the transformation being used to align the second input with the first.
Reset: Reset the current transformation values to their default values.
Display
Show Ground: When enabled, the ground plane will be displayed in the Cinema window.
Show Horizon: When enabled, the horizon line will be displayed in the Cinema window.
Show Names: When enabled, selected trackers will have their name displayed in the Cinema window
Show Info: When enabled, selected trackers will have positional information displayed in the Cinema window.
Trackers
The tracker list displays trackers provided by each input. A tracker will only be enabled in the list if it is present in both inputs. Trackers present in both inputs can be activated or deactivated. Only active trackers will be used during the automatic alignment process. The tracker list also displays the error for each tracker, which is the difference in the tracker position in both clips. Ideally, the error for each tracker should be close to zero after alignment.
All / None: Select all or none of the trackers from the list.
Activate: Activate all selected trackers. Active trackers will influence the automatic alignment process.
Deactivate: Deactivate all selected trackers. Inactive trackers will not influence the automatic alignment process.
Marquee: When enabled, the left mouse button can be used to draw a selection marquee in the Cinema window to select trackers. Pressing the Control key whilst drawing will ensure that previous selections are kept. Pressing the Shift key will allow a lasso selection to be used instead of a rectangle.
Orientation
Transform node: Specify a manual transformation mode for alignment. Translate, Rotate and Scale options will display a transform widget that can be edited using the left mouse button. Note that when Transform origin is set to Camera, the widget will only be visible in the Viewer windows. When in Fly mode, holding the Option key whist clicking and dragging in the a viewer window using the left, middle or right mouse buttons will translate, rotate and scale the secondary input.
Transform origin: This allows a choice between using the camera position, ground plane or the first selected tracker as the origin for manual transformations.
Auto Align: Automatically estimate a transformation to align trackers and frames that are common to both inputs into the coordinate system of the first input.
Refine: Refine the current transformation to better fit trackers and frames that are common to both inputs. This can be useful if automatic alignment fails, because an approximate transformation can be generated manually and then refined to produce a better fit.
Default Keyboard Shortcuts
Reset | Shift+E |
Translate | Shift+T |
Rotate | Shift+R |
Scale | Shift+S |
Fly | Shift+F |
Camera Origin | Shift+C |
Ground Origin | Shift+G |
Tracker Origin | Shift+Y |
Auto Align | Control+A |
Refine | Control+R |
Show Ground | Control+G |
Show Horizon | Control+H |
Show Names | Control+N |
Show Info | Control+I |
Next Clip | C |