Segment Anything
Alexander Kirillov, Eric Mintun, Nikhila Ravi et al. · IEEE/CVF International Conference on Computer Vision (ICCV) · 2023
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 segment-anything.com to foster research into foundation models for computer vision. We recommend reading the full paper at: arxiv.org/abs/2304.02643.
Cited by 10,910 works. doi:10.1109/iccv51070.2023.00371
Authors
- Eric Mintun
- Nikhila Ravi
- Hanzi Mao
- Chloe Rolland
- Laura Gustafson
- Tete Xiao
- Spencer Whitehead
- Alexander C. Berg
- Wan‐Yen Lo
- Piotr Dollár
- Ross Girshick
References
- Deep Residual Learning for Image Recognition
- Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
- A Computational Approach to Edge Detection
- Microsoft COCO: Common Objects in Context
- Snakes: Active contour models
- Focal Loss for Dense Object Detection
- Adaptive background mixture models for real-time tracking
- A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics
- Efficient Graph-Based Image Segmentation