INSID3: Training-Free In-Context Segmentation With DINOv3
MCML Authors
Christoph Reich
Daniel Cremers
Prof. Dr.
Director
Abstract
Christoph Reich
Daniel Cremers
Prof. Dr.
Director
Abstract
In-context segmentation (ICS) aims to segment arbitrary concepts, objects, parts, or personalized instances given a few annotated visual examples. Existing work relies on (i) fine-tuning vision foundation models (VFMs), which improves in-domain results but limits 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 concept 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 +6.1 % mIoU, while using 3x fewer parameters and without any mask or category-level supervision.
inproceedings CTR+26a
VisCon @CVPR 2026
3rd Workshop on Visual Concepts at the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Denver, CO, USA, Jun 03-07, 2026. Oral Presentation. To be published. Preprint available.Authors
C. Cuttano • G. Trivigno • C. Reich • D. Cremers • C. Masone • S. RothLinks
URLResearch Area
BibTeXKey: CTR+26a