normals = (normals - 90.) / 180. in saic_depth_completion/data/datasets/matterport.py

[ChristinaTan0704]

Hi, could you please explain a bit more about why we need to normalize the "normals" like this? If I understand correctly, the est_normal map from yindaz is ranging from 0~255 as they have: -- image.save(string.format("%s%s_normal_est.png", config.result_path, saveName), normal_outputs:add(1):mul(0.5)) https://github.com/yindaz/DeepCompletionRelease/blob/922fc064bb8f480be1d8b31a1b8755eaea8fb268/torch/main_test_matterport.lua#L71 Thanks!