Yongchang Zhang comments

Repositories
Issues
Comments

Results 4 comments of


                                            Yongchang Zhang

The detection score you provide is not normalized?

using sigmoid in trajectory 252,254

we can change **embedding_i = embedding_b[seg_mask_i] ** to **embedding_i = embedding_b * seg_mask_i**. but, the code **mean_i = torch.mean(embedding_i, dim=0)** means ? we should change ``` seg_mask_i = (seg_gt_b ==...

Discriminative Loss can not calculate dist loss...

@IrohXu what's means for ``` for j in range(0, embed_dim): seg_gt_b += seg_gt[b][j] * (j+1) ``` in your code?

Yongchang Zhang

The detection score you provide is not normalized?

linux20.04下出现无法分配颜色 none

[loss.py] there is a bug

Discriminative Loss can not calculate dist loss...