Segment Anything
This provides a foundational tool for computer vision researchers and practitioners by enabling flexible, zero-shot segmentation across diverse tasks and image distributions.
The Segment Anything project tackles the problem of creating a general-purpose image segmentation system by introducing a new promptable model and building the largest segmentation dataset to date with over 1 billion masks on 11 million images. The model achieves impressive zero-shot performance that is often competitive with or superior to prior fully supervised methods.
We introduce the Segment Anything (SA) project: a new task, model, and dataset for image segmentation. Using our efficient model in a data collection loop, we built the largest segmentation dataset to date (by far), with over 1 billion masks on 11M licensed and privacy respecting images. The model is designed and trained to be promptable, so it can transfer zero-shot to new image distributions and tasks. We evaluate its capabilities on numerous tasks and find that its zero-shot performance is impressive -- often competitive with or even superior to prior fully supervised results. We are releasing the Segment Anything Model (SAM) and corresponding dataset (SA-1B) of 1B masks and 11M images at https://segment-anything.com to foster research into foundation models for computer vision.