CMaP-SAM: Contraction Mapping Prior for SAM-driven Few-shot Segmentation

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Chen, Shuai, Meng, Fanman, Lei, Liming, Wei, Haoran, Wu, Chenhao, Wu, Qingbo, Xu, Linfeng, Li, Hongliang
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866912653272678400
author Chen, Shuai
Meng, Fanman
Lei, Liming
Wei, Haoran
Wu, Chenhao
Wu, Qingbo
Xu, Linfeng
Li, Hongliang
author_facet Chen, Shuai
Meng, Fanman
Lei, Liming
Wei, Haoran
Wu, Chenhao
Wu, Qingbo
Xu, Linfeng
Li, Hongliang
contents Few-shot segmentation (FSS) aims to segment new classes using few annotated images. While recent FSS methods have shown considerable improvements by leveraging Segment Anything Model (SAM), they face two critical limitations: insufficient utilization of structural correlations in query images, and significant information loss when converting continuous position priors to discrete point prompts. To address these challenges, we propose CMaP-SAM, a novel framework that introduces contraction mapping theory to optimize position priors for SAM-driven few-shot segmentation. CMaP-SAM consists of three key components: (1) a contraction mapping module that formulates position prior optimization as a Banach contraction mapping with convergence guarantees. This module iteratively refines position priors through pixel-wise structural similarity, generating a converged prior that preserves both semantic guidance from reference images and structural correlations in query images; (2) an adaptive distribution alignment module bridging continuous priors with SAM's binary mask prompt encoder; and (3) a foreground-background decoupled refinement architecture producing accurate final segmentation masks. Extensive experiments demonstrate CMaP-SAM's effectiveness, achieving state-of-the-art performance with 71.1 mIoU on PASCAL-$5^i$ and 56.1 on COCO-$20^i$ datasets. Code is available at https://github.com/Chenfan0206/CMaP-SAM.
format Preprint
id arxiv_https___arxiv_org_abs_2504_05049
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle CMaP-SAM: Contraction Mapping Prior for SAM-driven Few-shot Segmentation
Chen, Shuai
Meng, Fanman
Lei, Liming
Wei, Haoran
Wu, Chenhao
Wu, Qingbo
Xu, Linfeng
Li, Hongliang
Computer Vision and Pattern Recognition
Few-shot segmentation (FSS) aims to segment new classes using few annotated images. While recent FSS methods have shown considerable improvements by leveraging Segment Anything Model (SAM), they face two critical limitations: insufficient utilization of structural correlations in query images, and significant information loss when converting continuous position priors to discrete point prompts. To address these challenges, we propose CMaP-SAM, a novel framework that introduces contraction mapping theory to optimize position priors for SAM-driven few-shot segmentation. CMaP-SAM consists of three key components: (1) a contraction mapping module that formulates position prior optimization as a Banach contraction mapping with convergence guarantees. This module iteratively refines position priors through pixel-wise structural similarity, generating a converged prior that preserves both semantic guidance from reference images and structural correlations in query images; (2) an adaptive distribution alignment module bridging continuous priors with SAM's binary mask prompt encoder; and (3) a foreground-background decoupled refinement architecture producing accurate final segmentation masks. Extensive experiments demonstrate CMaP-SAM's effectiveness, achieving state-of-the-art performance with 71.1 mIoU on PASCAL-$5^i$ and 56.1 on COCO-$20^i$ datasets. Code is available at https://github.com/Chenfan0206/CMaP-SAM.
title CMaP-SAM: Contraction Mapping Prior for SAM-driven Few-shot Segmentation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2504.05049