Hang Chen

During inference, a "dummy" ground-truth json file is needed. The json file is to provide information about the image (for example, the mapping from image id to image file name)...

Note that our method is for refinement instead of directly performing segmentation, therefore an instance segmentation network is needed to provide coarse results. Assuming you are using mmdetection, you may...

mmdetection has changed this link. Now you can refer to https://github.com/open-mmlab/mmdetection/issues/2693 or its document for details. You can also get an example here https://cloud.tsinghua.edu.cn/f/d86b4a4e67544dc494ae/?dl=1

I think you are misunderstanding the usage of BPR. BPR is a post-processing method instead of an instance segmentation model, which means it is used to **refine** the segmentation results...

I see, the 'bbox' and 'score' are not necessary.

> > > > > mmdetection has changed this link. Now you can refer to [open-mmlab/mmdetection#2693](https://github.com/open-mmlab/mmdetection/issues/2693) or its document for details. You can also get an example here https://cloud.tsinghua.edu.cn/f/d86b4a4e67544dc494ae/?dl=1 >...