pointnet.pytorch
pointnet.pytorch copied to clipboard
How T-Net is initialized as identity matrix?
According to the https://arxiv.org/abs/1612.00593 paper, the T-Net is initialized as identity matrix.
I fail to understand how this is the case when looking into the STNkd code.
If I understand correctly:
iden = Variable(torch.from_numpy(np.eye(self.k).flatten().astype(np.float32))).view(1,self.k*self.k).repeat(batchsize,1)
if x.is_cuda:
iden = iden.cuda()
x = x + iden
x = x.view(-1, self.k, self.k)
return x
This snippet adds the identity matrix to the regressed matrix. How the identity initialization is achieved?