sysuzyq issues

Repositories
Issues
Comments

Results 2 issues of


                                            sysuzyq

The dimensionality of the last conv's output is 6630，but how can we reshape it to 1470。

layer { name: "conv_reg" type: "Convolution" bottom: "add_conv2" top: "conv_reg" param { lr_mult: 1 decay_mult: 1 } param { lr_mult: 2 decay_mult: 0 } convolution_param { num_output: 30 ################output :6...

Lowe’s first-to-second neighbor ratio test

nice work. https://github.com/naver/fire/blob/main/losses.py#L46 ``` dist2_second2 = torch.argmin(dist2, dim=0) ratio1to2 = dist[best2,arange] / dist2_second2. # should it be dist[best2,arange] /dist2[dist2_second2:arange] ? ```

sysuzyq

The dimensionality of the last conv's output is 6*6*30，but how can we reshape it to 1470。

Lowe’s first-to-second neighbor ratio test

The dimensionality of the last conv's output is 6630，but how can we reshape it to 1470。