Open shuaiyongjiang opened 1 year ago
Thanks for pointing out this issue. This is our mistake. "query_out = self.conv1(query_att + query)" should be "query_out = self.conv2(query_att + query)".
This mistake is an engineering code error and will not affect our motivation to design the network. Thanks again!
In Channel-wise Correlation Module(line 97):
two "self.conv1" are sharing weight?