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

[Fix] type of "counts" in COCO’s compressed RLE #9274

Merged
merged 2 commits into from
Nov 15, 2022
Merged

[Fix] type of "counts" in COCO’s compressed RLE #9274

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
6bd0ae9
[Fix]: fix wrong padding value in cached MixUp (#9259)
plutoyuxie Nov 8, 2022
1dde6b6
[Fix] type of "counts" in COCO’s compressed RLE
Daa98 Nov 8, 2022
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
2 changes: 1 addition & 1 deletion mmdet/datasets/transforms/loading.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -348,7 +348,7 @@ def _process_masks(self, results: dict) -> list:
elif isinstance(gt_mask, dict) and \
not (gt_mask.get('counts') is not None and
gt_mask.get('size') is not None and
isinstance(gt_mask['counts'], list)):
isinstance(gt_mask['counts'], (list, str))):
BIGWangYuDong marked this conversation as resolved.
Show resolved Hide resolved
# if gt_mask is a dict, it should include `counts` and `size`,
# so that `BitmapMasks` can uncompressed RLE
instance['ignore_flag'] = 1
Expand Down
12 changes: 6 additions & 6 deletions mmdet/datasets/transforms/transforms.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2464,9 +2464,9 @@ def transform(self, results: dict) -> dict:
ori_img = results['img']
origin_h, origin_w = out_img.shape[:2]
target_h, target_w = ori_img.shape[:2]
padded_img = np.zeros(
(max(origin_h, target_h), max(origin_w,
target_w), 3)).astype(np.uint8)
padded_img = np.ones((max(origin_h, target_h), max(
origin_w, target_w), 3)) * self.pad_val
padded_img = padded_img.astype(np.uint8)
padded_img[:origin_h, :origin_w] = out_img

x_offset, y_offset = 0, 0
Expand Down Expand Up @@ -3512,9 +3512,9 @@ def transform(self, results: dict) -> dict:
ori_img = results['img']
origin_h, origin_w = out_img.shape[:2]
target_h, target_w = ori_img.shape[:2]
padded_img = np.zeros(
(max(origin_h, target_h), max(origin_w,
target_w), 3)).astype(np.uint8)
padded_img = np.ones((max(origin_h, target_h), max(
origin_w, target_w), 3)) * self.pad_val
padded_img = padded_img.astype(np.uint8)
padded_img[:origin_h, :origin_w] = out_img

x_offset, y_offset = 0, 0
Expand Down