Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.28480 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911553636270080 |
|---|---|
| author | Cuttano, Claudia Trivigno, Gabriele Reich, Christoph Cremers, Daniel Masone, Carlo Roth, Stefan |
| author_facet | Cuttano, Claudia Trivigno, Gabriele Reich, Christoph Cremers, Daniel Masone, Carlo Roth, Stefan |
| contents | In-context segmentation (ICS) aims to segment arbitrary concepts, e.g., objects, parts, or personalized instances, given one annotated visual examples. Existing work relies on (i) fine-tuning vision foundation models (VFMs), which improves in-domain results but harms generalization, or (ii) combines multiple frozen VFMs, which preserves generalization but yields architectural complexity and fixed segmentation granularities. We revisit ICS from a minimalist perspective and ask: Can a single self-supervised backbone support both semantic matching and segmentation, without any supervision or auxiliary models? We show that scaled-up dense self-supervised features from DINOv3 exhibit strong spatial structure and semantic correspondence. We introduce INSID3, a training-free approach that segments concepts at varying granularities only from frozen DINOv3 features, given an in-context example. INSID3 achieves state-of-the-art results across one-shot semantic, part, and personalized segmentation, outperforming previous work by +7.5 % mIoU, while using 3x fewer parameters and without any mask or category-level supervision. Code is available at https://github.com/visinf/INSID3 . |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_28480 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | INSID3: Training-Free In-Context Segmentation with DINOv3 Cuttano, Claudia Trivigno, Gabriele Reich, Christoph Cremers, Daniel Masone, Carlo Roth, Stefan Computer Vision and Pattern Recognition In-context segmentation (ICS) aims to segment arbitrary concepts, e.g., objects, parts, or personalized instances, given one annotated visual examples. Existing work relies on (i) fine-tuning vision foundation models (VFMs), which improves in-domain results but harms generalization, or (ii) combines multiple frozen VFMs, which preserves generalization but yields architectural complexity and fixed segmentation granularities. We revisit ICS from a minimalist perspective and ask: Can a single self-supervised backbone support both semantic matching and segmentation, without any supervision or auxiliary models? We show that scaled-up dense self-supervised features from DINOv3 exhibit strong spatial structure and semantic correspondence. We introduce INSID3, a training-free approach that segments concepts at varying granularities only from frozen DINOv3 features, given an in-context example. INSID3 achieves state-of-the-art results across one-shot semantic, part, and personalized segmentation, outperforming previous work by +7.5 % mIoU, while using 3x fewer parameters and without any mask or category-level supervision. Code is available at https://github.com/visinf/INSID3 . |
| title | INSID3: Training-Free In-Context Segmentation with DINOv3 |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.28480 |