How to use depth maps?

[ycyentropy]

I have achieved real-time inference function by connecting the Realsense RGB-D camera, but the current model input only uses RGB maps. Is there any way to use depth maps to optimize human pose estimation?

[wqyin]

We are pleased to hear that you have achieved real-time inference. However, our work focuses exclusively on pose estimation from single-view RGB images.

[hhebb]

@ycyentropy
hi
could you let me know how did you do? thanks