Robust Conditional 3D Shape Generation from Casual Captures

7 hours ago (facebookresearch.github.io)

Does this need depth data capture as well? The “casual captures” makes it seem like it only needs images, but apparently they are using depth data as well

Also, can it run on Apple silicon?

  • I think it does use depth data from parameters in docs: python infer_shape.py --input_pkl <sample.pkl> (possibly achievable using software like MapAnything). I believe CUDA only.