Skip to main content

> ML_LITERATURE // KIRILLOV-2023-SEGMENT-ANYTHING-SAM_v1.0

Segment Anything (SAM)

Alexander Kirillov, Eric Mintun, Nikhila Ravi, Hanzi Mao, Chloe Rolland, Laura Gustafson, Tete Xiao, Spencer Whitehead, Alexander C. Berg, Wan-Yen Lo, Piotr Dollár, Ross Girshick · IEEE International Conference on Computer Vision (ICCV) (2023)

seminal-architecture2023industry-standardthirdPartyReproduced
DOI: 10.1109/ICCV51070.2023.00371Back to Literature

Principal Contribution

Introduced the Segment Anything Model (SAM) and SA-1B dataset (1 billion masks), establishing promptable zero-shot general visual segmentation.

Operational Relevance

Serves as canonical technical reference for implementing task-image-segmentation, task-zero-shot-classification in production systems.

Assumptions

  • Underlying data distribution satisfies empirical consistency and regularity assumptions across training domains

Limitations

  • Scaling characteristics and accuracy depend on hardware architecture, parameter scale, and dataset quality

Connected Algorithms, Architectures & Tools

Related Algorithms:
alg-self-distillation-dino
Related Architectures:
Implementing Libraries: