dgcnn.pytorch icon indicating copy to clipboard operation
dgcnn.pytorch copied to clipboard

Why the implemented network architecture is different with the architecture in the paper?

Open linlinge opened this issue 2 years ago • 1 comments

Hello! I have noticed a discrepancy between the network architecture used in the implemented code and the architecture described in the original paper. Below, you can see a figure depicting the architecture of the class DGCNN_semseg_s3dis, as plotted by me. Do I make any mistakes, or if the implemented code deviates from the architecture outlined in the original paper?

DGCNN

linlinge avatar Aug 21 '23 12:08 linlinge

I follow the original TensorFlow code in here. The architecture in this repo is the same as the original code.

antao97 avatar Aug 25 '23 02:08 antao97