Jin Zeng

icon indicating a website link https://jzengust.github.io/

Results 2 comments of Jin Zeng

would it be possible to replace gradient(calculated from autograd function) backpropagation by scalar loss backpropagation?

Hi Alisa, backward(df) is intended to zero out the gradient of pixels in the masked region. I didn't really try to use the depth branch alone for raw depth refinement,...

Python or C++ Implementation

Hi, Sorry, but we don't have python or c++ version of the code. If you have any questions regarding the implementation, please let me know. Thanks, Jin