Is SAM3 ready for pathology segmentation?

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kong, Qiuyu, Sharifi, Shakiba, Wang, Yiming, Cristani, Marco, Ruan, Zanxi
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909054787387392
author Kong, Qiuyu
Sharifi, Shakiba
Wang, Yiming
Cristani, Marco
Ruan, Zanxi
author_facet Kong, Qiuyu
Sharifi, Shakiba
Wang, Yiming
Cristani, Marco
Ruan, Zanxi
contents Is Segment Anything Model 3 (SAM3) capable in segmenting Any Pathology Images? Digital pathology segmentation spans tissue-level and nuclei-level scales, where traditional methods often suffer from high annotation costs and poor generalization. SAM3 introduces Promptable Concept Segmentation, offering a potential automated interface via text prompts. With this work, we propose a systematic evaluation protocol to explore the capability space of SAM3 in a structured manner. Specifically, we evaluate SAM3 under different supervision settings including zero-shot, few-shot, and supervised with varying prompting strategies. Our extensive evaluation on pathological datasets including NuInsSeg, PanNuke and GlaS, reveals that: (1) text-only prompts poorly activate nuclear concepts; (2) performance is highly sensitive to visual prompt types and budgets; (3) few-shot learning offers gains, but SAM3 lacks robustness against visual prompt noise; and (4) a significant gap persists between prompt-based usage and task-trained adapter-based reference. Our study delineates SAM3's boundaries in pathology image segmentation and provides practical guidance on the necessity of pathology domain adaptation.
format Preprint
id arxiv_https___arxiv_org_abs_2604_18225
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Is SAM3 ready for pathology segmentation?
Kong, Qiuyu
Sharifi, Shakiba
Wang, Yiming
Cristani, Marco
Ruan, Zanxi
Computer Vision and Pattern Recognition
Artificial Intelligence
Is Segment Anything Model 3 (SAM3) capable in segmenting Any Pathology Images? Digital pathology segmentation spans tissue-level and nuclei-level scales, where traditional methods often suffer from high annotation costs and poor generalization. SAM3 introduces Promptable Concept Segmentation, offering a potential automated interface via text prompts. With this work, we propose a systematic evaluation protocol to explore the capability space of SAM3 in a structured manner. Specifically, we evaluate SAM3 under different supervision settings including zero-shot, few-shot, and supervised with varying prompting strategies. Our extensive evaluation on pathological datasets including NuInsSeg, PanNuke and GlaS, reveals that: (1) text-only prompts poorly activate nuclear concepts; (2) performance is highly sensitive to visual prompt types and budgets; (3) few-shot learning offers gains, but SAM3 lacks robustness against visual prompt noise; and (4) a significant gap persists between prompt-based usage and task-trained adapter-based reference. Our study delineates SAM3's boundaries in pathology image segmentation and provides practical guidance on the necessity of pathology domain adaptation.
title Is SAM3 ready for pathology segmentation?
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2604.18225