Hi author, in the Bilateral_voting function, I am confused by code like the following
left =(torch.bmm(c_map[:,:,3].contiguous().view(-1,row,column),hori_translation.transpose(3,2).view(-1,column,column))).view(batch, classnum,row,column)
According to my current knowledge, I think it is calculating the part X{9-j}(x+a,y+b) according to equation (11), may I ask the author what exactly it is doing, I hope the author can tell me (due to my weak code ability, I hope the author can be as detailed as possible). I would be grateful if the author could give me some advice.
Hi author, in the Bilateral_voting function, I am confused by code like the following left =(torch.bmm(c_map[:,:,3].contiguous().view(-1,row,column),hori_translation.transpose(3,2).view(-1,column,column))).view(batch, classnum,row,column) According to my current knowledge, I think it is calculating the part X{9-j}(x+a,y+b) according to equation (11), may I ask the author what exactly it is doing, I hope the author can tell me (due to my weak code ability, I hope the author can be as detailed as possible). I would be grateful if the author could give me some advice.