This repository serves as a curated collection of papers, projects, and resources related to Meta's Segment Anything (SAM) model and its derivatives. It aims to track and summarize research progress and applications of SAM across various computer vision domains, benefiting researchers and developers exploring advanced segmentation techniques.

How It Works

The repository functions as a comprehensive index, categorizing and linking to a vast array of research papers, derivative projects, and applications built upon or inspired by the Segment Anything model. It covers diverse areas such as medical imaging, robotics, 3D reconstruction, and video analysis, providing a centralized hub for the SAM ecosystem.

Quick Start & Requirements

This repository is a curated list of resources, not a runnable software package. To use the underlying technologies, refer to individual project pages and codebases linked within the repository.

Highlighted Details

• Tracks foundational models like CLIP, OWL-ViT, Grounding DINO, and SAM itself.
• Features extensive lists of derivative papers and projects, including optimizations (MobileSAM, FastSAM), domain adaptations (Medical-SAM, FoodSAM), and novel applications (Inpaint Anything, Track-Anything).
• Covers a wide spectrum of tasks: image segmentation, video segmentation, medical imaging, inpainting, 3D segmentation, remote sensing, tracking, and more.
• Includes a JavaScript SDK (SAMJS) for front-end integration.

Maintenance & Community

The repository is actively maintained, with recent updates noted for new models like Segment Anything 2 and SaLIP. It encourages community contributions via Stars and Sharing.

Licensing & Compatibility

The repository itself is released under the MIT license. However, the licensing of individual projects and models linked within the repository will vary and must be checked separately.

Limitations & Caveats

This is a curated list and does not provide direct access to run or test the listed models. Users must navigate to individual project pages for installation, requirements, and usage details.