0.1.0

What's Changed

• Fix error for RGB in commit function by @constantinpape in #30
• Randomize colourmap to make it easier to see when objects are committed by @GenevieveBuckley in #33
• Update example scripts to include data download by @GenevieveBuckley in #35
• Implement tiled prediction for image encoder by @constantinpape in #34
• Develop instance segmentation with tiling by @constantinpape in #24
• Implement computing mask from points and add it to 3d annotator by @constantinpape in #46
• Refactor predictor initialization in segment_from_prompts by @constantinpape in #47
• Implement 3d auto segmentation by @constantinpape in #48
• Fix segment_from_mask(use_mask=True) for non-square masks by @constantinpape in #50
• Improve box extension functionality by @constantinpape in #53
• Add flag to return viewer in annotators by @constantinpape in #55
• Implement image series annotator by @constantinpape in #56
• Fix issue in tracking annotator by @constantinpape in #59
• Update Prompt Generator - First Point as Any FG Point by @anwai98 in #61
• Tunable instance seg by @constantinpape in #62
• Fix issue in tracking with points and set default projection mode to … by @constantinpape in #63
• Set correct version in setup.py by @constantinpape in #66
• Update Mac Installation Guide by @constantinpape in #74
• issue 40 add input data signature to embeddings file. by @paulhfu in #67
• Refactor the wrong file dialog gui by @constantinpape in #78
• Add sample data script and update the annotator examples by @constantinpape in #79
• Start preparations for doc by @constantinpape in #80
• Fix bug in 3d embedding precomputation (without zarr caching) by @constantinpape in #82
• Fix some issues raised by mypy by @constantinpape in #83
• 70 Create a GUI entrypoint by @paulhfu in #75
• Remove unused ndim from load_image_data by @constantinpape in #84
• Quit QApplication after entry gui is terminated by @paulhfu in #87
• Start adding type annotations to sam_annotator by @constantinpape in #86
• Add workflow for deploying the docs by @constantinpape in #88
• Build doc only on releae by @constantinpape in #89
• Implement TiledAutomaticMaskGenerator by @constantinpape in #90

New Contributors

• @GenevieveBuckley made their first contribution in #33
• @paulhfu made their first contribution in #67

Full Changelog: 0.0.2...0.1.0

Support for box prompts

This release introduces:

• Support for bounding box prompts, which provide better segmentation results than points in many cases.
• Interactive tracking now uses a better heuristic to propagate masks across time, leading to better automatic tracking results.
• and fixes several small bugs that have lead to missing intermediate frames / slices in tracking / 3d segmentation

What's Changed

• Expose model type as argument by @constantinpape in #12
• Convert inputs to uint8 by @constantinpape in #15
• Update Prompted Generator - Add Box Prompts by @anwai98 in #13
• Add test for the point prompt generator by @constantinpape in #16
• Add Condition for Background Points by @anwai98 in #18
• Implement instance segmentation with background by @constantinpape in #21
• Implement box prompts for all annotators by @constantinpape in #22
• New release by @constantinpape in #25
• Minor documentation changes by @ajinkya-kulkarni in #26
• Create README.md by @constantinpape in #27
• Fixed typo(s) by @ajinkya-kulkarni in #28

New Contributors

• @ajinkya-kulkarni made their first contribution in #26

Full Changelog: 0.0.1...0.0.2

Initial version

Initial version of SegmentAnything for Microscopy. Implements prototype napari applications for:

• interactive 2d segmentation
• interactive 3d segmentation
• interactive tracking