Saved in:
Bibliographic Details
Main Authors: Tejero, Javier Gamazo, Schmid, Moritz, Neila, Pablo Márquez, Zinkernagel, Martin S., Wolf, Sebastian, Sznitman, Raphael
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2501.06836
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909454586347520
author Tejero, Javier Gamazo
Schmid, Moritz
Neila, Pablo Márquez
Zinkernagel, Martin S.
Wolf, Sebastian
Sznitman, Raphael
author_facet Tejero, Javier Gamazo
Schmid, Moritz
Neila, Pablo Márquez
Zinkernagel, Martin S.
Wolf, Sebastian
Sznitman, Raphael
contents This paper addresses the domain adaptation challenge for semantic segmentation in medical imaging. Despite the impressive performance of recent foundational segmentation models like SAM on natural images, they struggle with medical domain images. Beyond this, recent approaches that perform end-to-end fine-tuning of models are simply not computationally tractable. To address this, we propose a novel SAM adapter approach that minimizes the number of trainable parameters while achieving comparable performances to full fine-tuning. The proposed SAM adapter is strategically placed in the mask decoder, offering excellent and broad generalization capabilities and improved segmentation across both fully supervised and test-time domain adaptation tasks. Extensive validation on four datasets showcases the adapter's efficacy, outperforming existing methods while training less than 1% of SAM's total parameters.
format Preprint
id arxiv_https___arxiv_org_abs_2501_06836
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SAM-DA: Decoder Adapter for Efficient Medical Domain Adaptation
Tejero, Javier Gamazo
Schmid, Moritz
Neila, Pablo Márquez
Zinkernagel, Martin S.
Wolf, Sebastian
Sznitman, Raphael
Computer Vision and Pattern Recognition
This paper addresses the domain adaptation challenge for semantic segmentation in medical imaging. Despite the impressive performance of recent foundational segmentation models like SAM on natural images, they struggle with medical domain images. Beyond this, recent approaches that perform end-to-end fine-tuning of models are simply not computationally tractable. To address this, we propose a novel SAM adapter approach that minimizes the number of trainable parameters while achieving comparable performances to full fine-tuning. The proposed SAM adapter is strategically placed in the mask decoder, offering excellent and broad generalization capabilities and improved segmentation across both fully supervised and test-time domain adaptation tasks. Extensive validation on four datasets showcases the adapter's efficacy, outperforming existing methods while training less than 1% of SAM's total parameters.
title SAM-DA: Decoder Adapter for Efficient Medical Domain Adaptation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2501.06836