InfoSAM: Fine-Tuning the Segment Anything Model from An Information-Theoretic Perspective

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Yuanhong, Yuan, Muyao, Zhang, Weizhan, Gong, Tieliang, Wen, Wen, Ying, Jiangyong, Shi, Weijie
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910982054346752
author Zhang, Yuanhong
Yuan, Muyao
Zhang, Weizhan
Gong, Tieliang
Wen, Wen
Ying, Jiangyong
Shi, Weijie
author_facet Zhang, Yuanhong
Yuan, Muyao
Zhang, Weizhan
Gong, Tieliang
Wen, Wen
Ying, Jiangyong
Shi, Weijie
contents The Segment Anything Model (SAM), a vision foundation model, exhibits impressive zero-shot capabilities in general tasks but struggles in specialized domains. Parameter-efficient fine-tuning (PEFT) is a promising approach to unleash the potential of SAM in novel scenarios. However, existing PEFT methods for SAM neglect the domain-invariant relations encoded in the pre-trained model. To bridge this gap, we propose InfoSAM, an information-theoretic approach that enhances SAM fine-tuning by distilling and preserving its pre-trained segmentation knowledge. Specifically, we formulate the knowledge transfer process as two novel mutual information-based objectives: (i) to compress the domain-invariant relation extracted from pre-trained SAM, excluding pseudo-invariant information as possible, and (ii) to maximize mutual information between the relational knowledge learned by the teacher (pre-trained SAM) and the student (fine-tuned model). The proposed InfoSAM establishes a robust distillation framework for PEFT of SAM. Extensive experiments across diverse benchmarks validate InfoSAM's effectiveness in improving SAM family's performance on real-world tasks, demonstrating its adaptability and superiority in handling specialized scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2505_21920
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle InfoSAM: Fine-Tuning the Segment Anything Model from An Information-Theoretic Perspective
Zhang, Yuanhong
Yuan, Muyao
Zhang, Weizhan
Gong, Tieliang
Wen, Wen
Ying, Jiangyong
Shi, Weijie
Computer Vision and Pattern Recognition
The Segment Anything Model (SAM), a vision foundation model, exhibits impressive zero-shot capabilities in general tasks but struggles in specialized domains. Parameter-efficient fine-tuning (PEFT) is a promising approach to unleash the potential of SAM in novel scenarios. However, existing PEFT methods for SAM neglect the domain-invariant relations encoded in the pre-trained model. To bridge this gap, we propose InfoSAM, an information-theoretic approach that enhances SAM fine-tuning by distilling and preserving its pre-trained segmentation knowledge. Specifically, we formulate the knowledge transfer process as two novel mutual information-based objectives: (i) to compress the domain-invariant relation extracted from pre-trained SAM, excluding pseudo-invariant information as possible, and (ii) to maximize mutual information between the relational knowledge learned by the teacher (pre-trained SAM) and the student (fine-tuned model). The proposed InfoSAM establishes a robust distillation framework for PEFT of SAM. Extensive experiments across diverse benchmarks validate InfoSAM's effectiveness in improving SAM family's performance on real-world tasks, demonstrating its adaptability and superiority in handling specialized scenarios.
title InfoSAM: Fine-Tuning the Segment Anything Model from An Information-Theoretic Perspective
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.21920