PixCap Docs
Video Mocap

Video mode

Video Mode uses AI to capture poses for every frame ranging from the given start and end frame of the chosen video.
The AI will perform automatic tracking of the person in the video sequence. Accurate tracking is not guaranteed (see more on quality and limitations below).
Output: An animation clip containing the full human poses of the video. The animation from this clip can be applied to any character in the PixCap scene that has been retargeted. Like motion capture data, the animation clip contains keyframes for each frame in the video sequence.

Quality and Limitations of Video Capture

For best possible results of Video Capture, the following are recommended:
  • Clear videos, and not too fast. The AI will not be able to perform well on very fast movements e.g. fast martial arts, gymnastics. Subtle movements such as hip-hop dancing and moonwalking will also not be captured.
  • Single actor in the video.
  • Full body visibility