SAM 3: Segment Anything with Concepts

By Nicolas Carion, Laura Gustafson, Yuan-Ting Hu, Shoubhik Debnath, Ronghang Hu, D'idac Sur'is, Chaitanya K. Ryali, Kalyan Vasudev Alwala, Haitham Khedr, Andrew Huang and 28 more. arXiv.org, 2025

636 citations recorded by the source corpus.

We present Segment Anything Model (SAM) 3, a unified model that detects, segments, and tracks objects in images and videos based on concept prompts, which we define as either short noun phrases (e.g.,"yellow school bus"), image exemplars, or a combination of both. Promptable Concept Segmentation (PCS) takes such prompts and returns segmentation masks and unique identities for all matching object instances. To advance PCS, we build a scalable data engine that produces a high-quality dataset with 4M unique concept labels, including hard negatives, across images and videos. Our model consists of an image-level detector and a memory-based video tracker that share a single backbone. Recognition and localization are decoupled with a presence head, which boosts detection accuracy. SAM 3 doubles the accuracy of existing systems in both image and video PCS, and improves previous SAM capabilities on visual segmentation tasks. We open source SAM 3 along with our new Segment Anything with Concepts (SA-Co) benchmark for promptable concept segmentation.

View Paper: https://www.semanticscholar.org/paper/9e341dd3dcf9ab98020625332700f84d56760c08

Citations in the corpus

• Low-cost concept-based localized explanations: How far can we get with training-free approaches? (2026, unknown): “The introduction of the Segment Anything Model (SAM) [13] decoupled localization from categorization, enabling modular pipelines where one component determines where a region is and another determines what it represents.”