FS-SAM2: Adapting Segment Anything Model 2 for Few-Shot Semantic Segmentation via Low-Rank Adaptation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Forni, Bernardo, Lombardi, Gabriele, Pozzi, Federico, Planamente, Mirco
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911155679657984
author Forni, Bernardo
Lombardi, Gabriele
Pozzi, Federico
Planamente, Mirco
author_facet Forni, Bernardo
Lombardi, Gabriele
Pozzi, Federico
Planamente, Mirco
contents Few-shot semantic segmentation has recently attracted great attention. The goal is to develop a model capable of segmenting unseen classes using only a few annotated samples. Most existing approaches adapt a pre-trained model by training from scratch an additional module. Achieving optimal performance with these approaches requires extensive training on large-scale datasets. The Segment Anything Model 2 (SAM2) is a foundational model for zero-shot image and video segmentation with a modular design. In this paper, we propose a Few-Shot segmentation method based on SAM2 (FS-SAM2), where SAM2's video capabilities are directly repurposed for the few-shot task. Moreover, we apply a Low-Rank Adaptation (LoRA) to the original modules in order to handle the diverse images typically found in standard datasets, unlike the temporally connected frames used in SAM2's pre-training. With this approach, only a small number of parameters is meta-trained, which effectively adapts SAM2 while benefiting from its impressive segmentation performance. Our method supports any K-shot configuration. We evaluate FS-SAM2 on the PASCAL-5$^i$, COCO-20$^i$ and FSS-1000 datasets, achieving remarkable results and demonstrating excellent computational efficiency during inference. Code is available at https://github.com/fornib/FS-SAM2
format Preprint
id arxiv_https___arxiv_org_abs_2509_12105
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FS-SAM2: Adapting Segment Anything Model 2 for Few-Shot Semantic Segmentation via Low-Rank Adaptation
Forni, Bernardo
Lombardi, Gabriele
Pozzi, Federico
Planamente, Mirco
Computer Vision and Pattern Recognition
Few-shot semantic segmentation has recently attracted great attention. The goal is to develop a model capable of segmenting unseen classes using only a few annotated samples. Most existing approaches adapt a pre-trained model by training from scratch an additional module. Achieving optimal performance with these approaches requires extensive training on large-scale datasets. The Segment Anything Model 2 (SAM2) is a foundational model for zero-shot image and video segmentation with a modular design. In this paper, we propose a Few-Shot segmentation method based on SAM2 (FS-SAM2), where SAM2's video capabilities are directly repurposed for the few-shot task. Moreover, we apply a Low-Rank Adaptation (LoRA) to the original modules in order to handle the diverse images typically found in standard datasets, unlike the temporally connected frames used in SAM2's pre-training. With this approach, only a small number of parameters is meta-trained, which effectively adapts SAM2 while benefiting from its impressive segmentation performance. Our method supports any K-shot configuration. We evaluate FS-SAM2 on the PASCAL-5$^i$, COCO-20$^i$ and FSS-1000 datasets, achieving remarkable results and demonstrating excellent computational efficiency during inference. Code is available at https://github.com/fornib/FS-SAM2
title FS-SAM2: Adapting Segment Anything Model 2 for Few-Shot Semantic Segmentation via Low-Rank Adaptation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.12105