Saved in:
Bibliographic Details
Main Authors: Yu, Meng, Cui, Te, Chu, Qitong, Song, Wenjie, Yang, Yi, Yue, Yufeng
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2506.21975
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915361401602048
author Yu, Meng
Cui, Te
Chu, Qitong
Song, Wenjie
Yang, Yi
Yue, Yufeng
author_facet Yu, Meng
Cui, Te
Chu, Qitong
Song, Wenjie
Yang, Yi
Yue, Yufeng
contents Reliable semantic segmentation of open environments is essential for intelligent systems, yet significant problems remain: 1) Existing RGB-T semantic segmentation models mainly rely on low-level visual features and lack high-level textual information, which struggle with accurate segmentation when categories share similar visual characteristics. 2) While SAM excels in instance-level segmentation, integrating it with thermal images and text is hindered by modality heterogeneity and computational inefficiency. To address these, we propose TASeg, a text-aware RGB-T segmentation framework by using Low-Rank Adaptation (LoRA) fine-tuning technology to adapt vision foundation models. Specifically, we propose a Dynamic Feature Fusion Module (DFFM) in the image encoder, which effectively merges features from multiple visual modalities while freezing SAM's original transformer blocks. Additionally, we incorporate CLIP-generated text embeddings in the mask decoder to enable semantic alignment, which further rectifies the classification error and improves the semantic understanding accuracy. Experimental results across diverse datasets demonstrate that our method achieves superior performance in challenging scenarios with fewer trainable parameters.
format Preprint
id arxiv_https___arxiv_org_abs_2506_21975
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle TASeg: Text-aware RGB-T Semantic Segmentation based on Fine-tuning Vision Foundation Models
Yu, Meng
Cui, Te
Chu, Qitong
Song, Wenjie
Yang, Yi
Yue, Yufeng
Computer Vision and Pattern Recognition
Reliable semantic segmentation of open environments is essential for intelligent systems, yet significant problems remain: 1) Existing RGB-T semantic segmentation models mainly rely on low-level visual features and lack high-level textual information, which struggle with accurate segmentation when categories share similar visual characteristics. 2) While SAM excels in instance-level segmentation, integrating it with thermal images and text is hindered by modality heterogeneity and computational inefficiency. To address these, we propose TASeg, a text-aware RGB-T segmentation framework by using Low-Rank Adaptation (LoRA) fine-tuning technology to adapt vision foundation models. Specifically, we propose a Dynamic Feature Fusion Module (DFFM) in the image encoder, which effectively merges features from multiple visual modalities while freezing SAM's original transformer blocks. Additionally, we incorporate CLIP-generated text embeddings in the mask decoder to enable semantic alignment, which further rectifies the classification error and improves the semantic understanding accuracy. Experimental results across diverse datasets demonstrate that our method achieves superior performance in challenging scenarios with fewer trainable parameters.
title TASeg: Text-aware RGB-T Semantic Segmentation based on Fine-tuning Vision Foundation Models
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.21975