gakkii
Thank you so much for pointing this out. I am also doing the kitti format conversion. Is there anything else I should pay attention to? I replaced most of the...
Oh, I can comment an assertion without modifying the axis definition, which seems to get the correct visualization, use the render module in the script. @bmankirLinker
@Tai-Wang Sorry, I don't know.
@pyaf Sorry for my belated clarification. I ignored the transformation of the coordinate system at that time, and in fact the BOX given is the same. I'm sorry very much.
Also wondering, is it possible to just change the output of rcnn_cls?
The visiulize code(demo part) has been updated.
This file(postprocessing.py in detectron2) needs to be replaced. ``` First replace the original detectron2 installed postprocessing.py with the file. ``` This is due to Detectron2's postprocessing.py only supports square mask...
@jinfagang Initially, this was a bug here https://github.com/gakkiri/SOLOv2-detectron2/blob/master/SOLOv2/modeling/solov2.py#L180. That's when I used .gt(0), causing Person to be ignored, and other categories are well trained. After I fixed the problem and...
@jinfagang No, all categories of AP are not high(about 10+-), but the losses converged to a good level. I think maybe the problem is somewhere else. Although at that time...
@jinfagang BTW, It works well when there's no Person, and that's what surprises me.