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

Allow flexible post-processing for larger segmentations #853

Merged
merged 3 commits into from
Feb 6, 2025
Merged

Allow flexible post-processing for larger segmentations #853

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
2882d70
Allow flexible post-processing for larger segmentations
anwai98 Feb 5, 2025
3dd0481
Add comments to code structure
anwai98 Feb 5, 2025
c9c46c6
Simplify logic to return empty masks
anwai98 Feb 6, 2025
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
17 changes: 8 additions & 9 deletions micro_sam/automatic_segmentation.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -127,17 +127,16 @@ def automatic_instance_segmentation(
segmenter.initialize(image=image_data, image_embeddings=image_embeddings, verbose=verbose)
masks = segmenter.generate(**generate_kwargs)

if len(masks) == 0: # instance segmentation can have no masks, hence we just save empty labels
if isinstance(segmenter, InstanceSegmentationWithDecoder):
this_shape = segmenter._foreground.shape
elif isinstance(segmenter, AMGBase):
this_shape = segmenter._original_size
if isinstance(masks, list):
# whether the predictions from 'generate' are list of dict,
# which contains additional info req. for post-processing, eg. area per object.
if len(masks) == 0:
instances = np.zeros(image_data.shape[-2:], dtype="uint32")
else:
this_shape = image_data.shape[-2:]

instances = np.zeros(this_shape, dtype="uint32")
instances = mask_data_to_segmentation(masks, with_background=True, min_object_size=0)
else:
instances = mask_data_to_segmentation(masks, with_background=True, min_object_size=0)
# if (raw) predictions provided, store them as it is w/o further post-processing.
instances = masks

else:
if (image_data.ndim != 3) and (image_data.ndim != 4 and image_data.shape[-1] != 3):
Expand Down