search

maudzung / Complex-YOLOv4-Pytorch

The PyTorch Implementation based on YOLOv4 of the paper: "Complex-YOLO: Real-time 3D Object Detection on Point Clouds"
https://arxiv.org/pdf/1803.06199.pdf
GNU General Public License v3.0
1.24k stars 256 forks source link

Lack of backword pass in Giou module #13

Closed chaomath closed 4 years ago

chaomath commented 4 years ago 
convex_conners = torch.cat((p_cons, t_cons), dim=0)
hull = ConvexHull(convex_conners.clone().detach().cpu().numpy())  # done on cpu, just need indices output
convex_conners = convex_conners[hull.vertices]
convex_polygon = cvt_box_2_polygon(convex_conners)
convex_area = convex_polygon.area
giou_loss += 1. - (iou - (convex_area - union) / (convex_area + 1e-16))

this problem can be here: https://github.com/facebookresearch/detectron2/issues/1347

maudzung commented 4 years ago

Thank you so much! I'm also trying to fix the bug these days.