SAM3-UNet: Simplified Adaptation of Segment Anything Model 3

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Xiong, Xinyu, Wu, Zihuang, Lu, Lei, Xia, Yufa
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911296441548800
author Xiong, Xinyu
Wu, Zihuang
Lu, Lei
Xia, Yufa
author_facet Xiong, Xinyu
Wu, Zihuang
Lu, Lei
Xia, Yufa
contents In this paper, we introduce SAM3-UNet, a simplified variant of Segment Anything Model 3 (SAM3), designed to adapt SAM3 for downstream tasks at a low cost. Our SAM3-UNet consists of three components: a SAM3 image encoder, a simple adapter for parameter-efficient fine-tuning, and a lightweight U-Net-style decoder. Preliminary experiments on multiple tasks, such as mirror detection and salient object detection, demonstrate that the proposed SAM3-UNet outperforms the prior SAM2-UNet and other state-of-the-art methods, while requiring less than 6 GB of GPU memory during training with a batch size of 12. The code is publicly available at https://github.com/WZH0120/SAM3-UNet.
format Preprint
id arxiv_https___arxiv_org_abs_2512_01789
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SAM3-UNet: Simplified Adaptation of Segment Anything Model 3
Xiong, Xinyu
Wu, Zihuang
Lu, Lei
Xia, Yufa
Computer Vision and Pattern Recognition
In this paper, we introduce SAM3-UNet, a simplified variant of Segment Anything Model 3 (SAM3), designed to adapt SAM3 for downstream tasks at a low cost. Our SAM3-UNet consists of three components: a SAM3 image encoder, a simple adapter for parameter-efficient fine-tuning, and a lightweight U-Net-style decoder. Preliminary experiments on multiple tasks, such as mirror detection and salient object detection, demonstrate that the proposed SAM3-UNet outperforms the prior SAM2-UNet and other state-of-the-art methods, while requiring less than 6 GB of GPU memory during training with a batch size of 12. The code is publicly available at https://github.com/WZH0120/SAM3-UNet.
title SAM3-UNet: Simplified Adaptation of Segment Anything Model 3
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.01789