S2A：面向无对齐RGB-T显著目标检测的语义到空间对齐方法

中文摘要 AI 辅助

无对齐RGB-T显著目标检测（RGB-T SOD）旨在从未配准的RGB和热红外图像对中识别显著目标，而无需昂贵的预对齐处理。然而，空间错位破坏了像素级对应关系，并在跨模态融合过程中导致特征污染。为解决这一问题，我们提出了S2A，一种用于无对齐RGB-T SOD的语义到空间对齐框架。具体而言，全局引导分层融合模块（GGHF）首先利用全局语义引导来抑制背景干扰并细化分层模态内特征。随后，无对齐跨模态通道注意力模块（AFCA）通过通道交互全局交换互补语义信息，有效克服局部空间错位造成的干扰。最后，空间可变形交叉注意力模块（SDCA）预测自适应采样偏移以恢复局部跨模态空间对应关系。通过这种语义到空间范式，S2A首先实现可靠的跨模态语义交互，随后进行局部空间校准，有效减少错位引起的特征污染。无需额外复杂设计，S2A在多个公开的无对齐RGB-T基准上取得了极具竞争力的性能，证明了其在缓解错位引起的特征污染方面的有效性。

英文摘要

Alignment-free RGB-T salient object detection (RGB-T SOD) aims to identify salient objects from unregistered RGB and thermal image pairs without costly pre-alignment. However, spatial misalignment breaks pixel-wise correspondence and causes feature contamination during cross-modal fusion. To address this issue, we propose S2A, a semantic-to-spatial alignment framework for alignment-free RGB-T SOD. Specifically, a global-guided hierarchical fusion module (GGHF) first exploits global semantic guidance to suppress background interference and refine hierarchical intra-modal features. Subsequently, the alignment-free cross-modal channel attention module (AFCA) globally exchanges complementary semantic information through channel-wise interaction, effectively overcoming the interference caused by local spatial misalignments. Finally, a spatial deformable cross-attention module (SDCA) predicts adaptive sampling offsets to recover local cross-modal spatial correspondence. Through this semantic-to-spatial paradigm, S2A first enables reliable cross-modal semantic interaction and subsequently performs local spatial calibration, effectively reducing misalignment-induced feature contamination. Without bells and whistles, S2A achieves highly competitive performance on multiple public alignment-free RGB-T benchmarks, demonstrating its effectiveness in alleviating misalignment-induced feature contamination.

发表机构

• Jiangxi Normal University（江西师范大学）

机构由 AI 辅助整理，请以论文原文为准。