Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[BUG FIX] Remove unnecessary padding during NMS #2584

Closed
wants to merge 2 commits into from
Closed

[BUG FIX] Remove unnecessary padding during NMS #2584

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
9c50411
Fix mask dimensions mismatch in rtmdet-ins models after exporting to …
LuukvandenBent Nov 29, 2023
2cbd3fe
Remove unnecessary padding of bounding box
LuukvandenBent Dec 1, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 1 addition & 5 deletions mmdeploy/mmcv/ops/nms.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -349,10 +349,6 @@ def _multiclass_nms_single(boxes: Tensor,
dets = torch.cat([boxes, scores], dim=2)
labels = cls_inds.unsqueeze(0)

# pad
dets = torch.cat((dets, dets.new_zeros((1, 1, 5))), 1)
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

this padding is useful for case when selected_indices is zero shaped tensor.

labels = torch.cat((labels, labels.new_zeros((1, 1))), 1)

# topk or sort
is_use_topk = keep_top_k > 0 and \
(torch.onnx.is_in_onnx_export() or keep_top_k < dets.shape[1])
Expand All @@ -369,7 +365,7 @@ def _multiclass_nms_single(boxes: Tensor,
if pre_top_k > 0:
bbox_index = pre_topk_inds[None, box_inds]
if keep_top_k > 0:
bbox_index = bbox_index[:, topk_inds[:-1]]
bbox_index = bbox_index[:, topk_inds]
return dets, labels, bbox_index
else:
return dets, labels
Expand Down
Loading