4 channel Input for RGB-D data

[advaitsave]

Hello,
I am working with a depth sensing camera (Intel® RealSense™ D415), which can produce RGB and Depth frames. I was hoping to train deepposekit models with RGBD 4 channel input, and was wondering if this is something anyone has tried.

Would it be feasible to modify the library to accept 4 channel inputs? Any guidance on how to do this would help!

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs.

[stale]

This issue has been automatically closed because it has not had recent activity.