DIoU-pytorch-detectron
DIoU-pytorch-detectron copied to clipboard
Zzh-tju
Distance-IoU Loss into Faster R-CNN
CIou
Hi, tju Thanks for your project. I try to use the ciou loss and add it to the project https://github.com/biubug6/Pytorch_Retinaface But it seem to get the same precision AP. Do...
shoule i change both roi's loss_bbox and rpn's loss_bbox from l1 loss to ciou loss ,and for doule shared head ?
Hello, Could you please explain why did you normalize the loss at the end by both: - bbox_outside_weights - output.size(0) ? > iou_weights = bbox_inside_weights.view(-1, 4).mean(1) * bbox_outside_weights.view(-1, 4).mean(1) >...
System information: Operating system: Ubuntu 20.04 CUDA version: 11.0 python version: 3.8 pytorch version: 1.6.0 sh make.sh > Traceback (most recent call last): File "build.py", line 4, in from torch.utils.ffi...
Got Nan
Hello, thanks for your code. Refering to your code, I reimplemented CIoU loss in my faster-rcnn framework, but got Nan gradients in the first iteration of training. Tracking the Nan...
I have a question for DIOU code. I use the same code as your repo https://github.com/Zzh-tju/DIoU-pytorch-detectron, while I cannot get the same result. Surprisingly, even when I use the IOU...
when sh make.sh there is this error. and pytorch version is ?
投的什么会议？没发表就放在arvix上？
I am using CIOU for training where there can be multiple boxes at once, and I am getting NaN by your compute_ciou() loss function.
