INSID3: Training-Free In-Context Segmentation with DINOv3

TL;DR

INSID3 is a training-free method that leverages frozen DINOv3 features for versatile in-context segmentation, achieving state-of-the-art results without supervision or auxiliary models.

Contribution

It demonstrates that a single self-supervised backbone can support both semantic matching and segmentation without additional training or supervision.

Findings

Outperforms previous methods by +7.5% mIoU in segmentation tasks.

Uses 3x fewer parameters than prior approaches.

Operates without any mask or category-level supervision.

Abstract

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 .