Closed cbenitez81 closed 4 years ago
I check your code and I couldn't understand the graph feature with respect to dgcnn paper. It turns out that the original authors used to have the bug in line 72 of your code
feature = torch.cat((feature, x), dim=3).permute(0, 3, 1, 2)
and it should be
feature = torch.cat((feature - x, x), dim=3).permute(0, 3, 1, 2)
Thanks for de code, I haven't tested, I'm still checking that it matches the theory
Thanks for pointing it out. Actually we tested both cases and didn't find critical differences.
I check your code and I couldn't understand the graph feature with respect to dgcnn paper. It turns out that the original authors used to have the bug in line 72 of your code
feature = torch.cat((feature, x), dim=3).permute(0, 3, 1, 2)
and it should befeature = torch.cat((feature - x, x), dim=3).permute(0, 3, 1, 2)
Thanks for de code, I haven't tested, I'm still checking that it matches the theory