hustvl/CrossVIS

Hello,What do these parameters (fpn_levels_0/1, gt_inds_0/1, inst_im_inds,inst_rel_ids) mean in crossvis.py

Alxx999 opened this issue · 2 comments

Hello,What do these parameters (fpn_levels_0/1, gt_inds_0/1, inst_im_inds,inst_rel_ids) mean in crossvis.py

Hi, @Alxx999! Thanks for your attention in our work.
fpn_levels and gt_inds indicates the corresponding fpn feature and ground truth for each predicted instance, respectively. You can try running CondInst to get details. As we sample a pair of frames from one video in a batch, we use 0 and 1 to identify two frames.
inst_im_inds and inst_rel_ids are used to find the instance relations between a pair of frames.

你好,@Alxx999!感谢您对我们工作的关注。 fpn_levelsgt_inds分别表示每个预测实例对应的 fpn 特征和ground truth。您可以尝试运行CondInst以获取详细信息。当我们从一个视频中批量采样一对帧时,我们使用 0 和 1 来识别两个帧。 inst_im_indsinst_rel_ids用于查找一对帧之间的实例关系。

OK!Thank you very much!