Using Multiple Cameras


I’m new to DeepLabCut. From the tutorials, I’ve seen motion capture data being produced from a single camera source, and from a single angle. I’m wondering, in 3D motion capture, is it not better to produce a 3D animation using data from several camera angles/sources?

If so, how might this be accomplished using DeepLabCut? Is there a part in the tutorials I am missing?

Thank you!

I’m also new to DeepLabCut. but I have read a paper named: Automated Rat Single-Pellet Reaching with 3-Dimensional Reconstruction of Paw and Digit Trajectories. URL:; DOI: doi:10.3791/59979
This may help you!

1 Like

DeepLabCut also has 3D tools built into the package, please see the docs and video tutorials: