Detect Blobs [Active Range]

Uses the settings from the Blob Parameters section of the Properties pane to find blobs in every frame in the active range of the sequence to create detections. The detections are not joined between frames (i.e. there can be a detection for the same marker on frames 0 and 1, but this option does not connect them). For more information on how blob detection works, see Blob detection algorithm.

This process can be quite time-consuming so before running it, ensure you have the correct settings by enabling Blob Preview and scrubbing though the take to check the detections.

Note: Using Detect Blobs [Active Range] does not enable you to spot gaps in the tracking using the Continuity or Selected Point panes.

Detect Blobs [Current Frame]

Uses the settings from the Blob Parameters section of the Properties pane to find blobs in the current frame in the sequence to create detections. For more information on how blob detection works, see Blob detection algorithm.

This option is useful for detecting blobs in the first frame prior to initializing the 3D points for the frame, before tracking the entire take.

Load ROM

Loads a .cara file (normally a ROM file) that can be used to initialize the current take. For more information, see Initialize from a ROM file in Using Vicon CaraPost and also Initialize from ROM algorithm.

Initialize from ROM

Initializes the current take using the data (the face as well as the brim) from a .cara file that has been loaded using the Load ROM option (see above), using the parameters set in the Initialize From ROM Parameters section of the Properties pane. For more information, see Initialize from a ROM file in Using Vicon CaraPost and also Initialize from ROM algorithm.

Initialize from ROM [Brim Only]

Initializes the current take using the data (the brim only) from a .cara file that has been loaded using the Load ROM option (see above), using the parameters set in the Initialize From ROM Parameters section of the Properties pane. For more information, see Initialize from a ROM file in Using Vicon CaraPost and also Initialize from ROM algorithm.

Initialize Default Calibration Brim [v2.0]

Loads a 3D model of the markers on the current version (V2.0) of the default calibration brim.

Important: The default calibration brim is an object that is used by Vicon CaraPost to calculate the positions of the cameras on a per-frame basis. To enable CaraPost to do this correctly, it must use the correct calibration brim, If you are using a custom calibration brim, you must import it using the Import Calibration Brim [svg] option on the File menu instead.

Initialize Default Calibration Brim [v1.1]

Loads a 3D model of the markers on V1.1 of the default calibration brim.

Important: The default calibration brim is an object that is used by Vicon CaraPost to calculate the positions of the cameras on a per-frame basis. To enable CaraPost to do this correctly, it must use the correct calibration brim, If you are using a custom calibration brim you must import it using the Import Calibration Brim [svg] option on the File menu instead.

Initialize Default Calibration Brim [v1.0]

Loads a 3D model of the markers on V1.0 of the default calibration brim.

Important: The default calibration brim is an object that is used by Vicon CaraPost to calculate the positions of the cameras on a per-frame basis. To enable CaraPost to do this correctly, it must use the correct calibration brim, If you are using a custom calibration brim you must import it using the Import Calibration Brim [svg] option on the File menu instead.

Solve Calibration Brim [Manual]

If you have a camera calibration loaded, have initialized the correct calibration brim, and connected four or more detections to the calibration brim, this process calculates the camera positions based on the available information.

For more information, see Solve calibration brim algorithm.

Initialize 3D

Takes the current camera positions and tries to connect detections in different camera views and calculate 3D positions for them, using the settings defined in the Initialize 3D Parameters section of the Properties pane. To enable this process to work well, you must have reasonably accurate camera positions, as well as a number of hand-matched ‘seed’ points. The camera positions and ‘seed’ marker points are normally derived from solving the calibration brim, manually adding blob correspondences to give the ‘seed’ points, and finally, using a Bundle Unconstrained option (see below).

For more information, see 3D initialization algorithm.

Bundle Current Frame [Unconstrained]

Adjusts the 3D positions of all marker points so that the difference between each reprojected marker point and the center of the detection it is based on (re-projection error) is minimized across all camera views. Acts on the current frame only and adjusts the camera positions to minimize the re-projection error.

For more information, see Bundle adjustment algorithm.

Bundle Sequence [Unconstrained]

Adjusts the 3D positions of all marker points so that the difference between each reprojected marker point and the center of the detection they are based on (re-projection error) is minimized across all camera views. Acts on the entire sequence and adjusts the camera positions to minimize the reprojection error.

For more information, see Bundle adjustment algorithm.

Track Forwards [Sequence]

Takes the points in the current frame and tracks them for the rest of the take by matching detections between frames, using the settings defined in the Tracking Parameters section of the Properties pane. This option also calculates the 3D positions of cameras and points for every frame, and performs blob detection if it hasn't already been done.

For more information, see Tracking algorithm.

Track Backwards [Sequence]

Takes the points in the current frame and tracks them to the beginning of the take by matching detections between frames, using the settings defined in the Tracking Parameters section of the Properties pane. This option also calculates the 3D positions of cameras and points for every frame, and performs blob detection if it hasn't already been done.

For more information, see Tracking algorithm.

Track Forwards [Single Frame]

Takes the points in the current frame and tracks them into the next frame by matching detections between frames, using the settings defined in the Tracking Parameters section of the Properties pane. It also calculates the 3D positions of cameras and points for the next frame, performs blob detection if it hasn’t already been done, and aligns the calibration brim based on the previous frame.

For more information, see Tracking algorithm.

Track Backwards [Single Frame]

Takes the points in the current frame and tracks them into the previous frame by matching detections between frames, using the settings defined in the Tracking Parameters section of the Properties pane. It also calculates the 3D positions of cameras and points for the previous frame, performs blob detection if it hasn’t already been done, and aligns the calibration brim based on the current frame.

For more information, see Tracking algorithm.

Stabilize Data [Manual]

Takes a minimum of three gap-free points that you select as being relatively unaffected by changes in expression throughout the Active Range, and then for each frame maps those points on the current frame to the same points on the base frame. Vicon CaraPost reports the stabilization error at the end of the stabilization process.

For more information see Manual Stabilization in Using Vicon CaraPost. 

Stabilize Data [Assisted]

Like Stabilize Data [Manual] (see above), operates on a minimum of three gap-free points that you select. Normally produces better results than manual stabilization, as the assisted process enables you to customize the stabilization parameters to suit your particular take, and to benefit from Vicon CaraPost’s ability to contribute further stable points, based on your original selection, minimizing the stabilization error across the whole of the Active Range.

For more information see Assisted Stabilization in Using Vicon CaraPost. 

Align Data

Aligns the face data into a coordinate frame that you define by selecting three 3D face points.

Note: If you are stabilizing a take it is preferable to perform stabilization before you align the data. This is because if you stabilize after aligning data, the markers are returned to their original positions, as stabilization operates on the original, raw data, so you will need to repeat the Align Data operation.

For more information, see Align face data in 3D space in Using Vicon CaraPost.

Interpolate Gaps

After you have tracked and corrected a take, use this option to fill in any remaining gaps, using the settings defined in the Gap-Filling Parameters section of the Properties pane.

Fills in (or interpolates) gaps in the 3D point data, where either a 3D point is not present at all, or a point may be present but with only a detection connected in a single view. The function fills all gaps in the 3D information in the active range of the take for all selected points, or if no points are selected, for all points. Gap-filling is performed using a non-linear interpolation technique (see Gap interpolation algorithm), using the options selected in the Gap-Filling Parameters section of the Properties pane. If necessary, you can remove the interpolated data afterwards by selecting Remove Interpolated 3D Points on the Edit menu.