Closed prernadh closed 1 month ago
The recent updates enhance the FiftyOne framework by incorporating advanced image and video segmentation capabilities through the Segment Anything 2 (SAM2) model. Key additions include new classes for processing images and videos, improved input handling, and expanded testing functionalities. Additionally, refinements to existing components streamline the code's maintainability and flexibility, allowing for easier future enhancements.
Files | Change Summary |
---|---|
fiftyone/utils/sam.py , fiftyone/utils/sam2.py , tests/intensive/model_zoo_tests.py |
Introduced classes for image and video segmentation; improved input handling and testing framework for various prompt types. |
pylintrc |
Updated to include cv2 for better linting and dynamic member recognition; improved handling of OpenCV functionalities. |
In the meadow where pixels bloom,
A rabbit hops, dispelling gloom.
With SAM2's magic, we now can see,
Segments of joy, wild and free!
Hopping through frames, what a delight,
Image and video, all in sight! 🐰✨
Thank you for using CodeRabbit. We offer it for free to the OSS community and would appreciate your support in helping us grow. If you find it useful, would you consider giving us a shout-out on your favorite social media?
What changes are proposed in this pull request?
Adding Segment Anything 2 to the Fiftyone model zoo.
How is this patch tested? If it is not, please explain why.
Tested manually with different configurations -
Release Notes
Is this a user-facing change that should be mentioned in the release notes?
Added SAM2 into the Fiftyone model zoo with inference support for both images and videos.
What areas of FiftyOne does this PR affect?
fiftyone
Python library changesBox prompt for Images
Keypoint prompt for Images
Automatic segmentation for Images
Prompting for Videos
Summary by CodeRabbit
New Features
Improvements
Bug Fixes
Chores