面向遮挡可见光-红外行人重识别的空间-频率协同学习
Spatial-Frequency Collaborative Learning for Occluded Visible-Infrared Person Re-Identification
将频域特征作为空间特征的补充，通过跨模态频率对齐模块和空间-频率交互模块提升遮挡下的跨模态特征对齐能力，为可见光-红外行人重识别提供了一种可复用的双域协同学习思路。
原文依据 we introduce a spatial-frequency collaborative perspective that offers global perception and cross-location consistency
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论遮挡可见光-红外行人重识别（Occluded VI-ReID）由于模态异质性和遮挡而仍然困难，这两者都破坏了结构一致性并削弱了跨模态特征对齐。现有方法主要依赖空间域线索（如局部身体部位和显著块），但在变化的成像条件或部分可见性下，其判别力严重下降。为解决这些问题，我们引入了一种空间-频率协同视角，提供全局感知和跨位置一致性。具体而言，我们提出了一个空间-频率协同学习（SFCL）框架，利用频率信息来补充空间表示。SFCL 包括一个跨模态频率对齐模块（CFAM）、一个空间-频率交互模块（SFIM）和一个频率感知判别（FAD）损失。CFAM 在频域中对可见光/红外图像的光谱特征进行建模，建立模态一致的光谱先验。SFIM 将这些先验注入空间特征，促进双域交互以及空间和频率语义的互补表示。此外，FAD 损失联合强制跨模态频率对齐和语义一致性，从而增强遮挡下的鲁棒性和判别力。为了进行真实遮挡评估，我们构建了两个遮挡数据集 Occ-SYSU-MM01 和 Occ-RegDB，SFCL 在这些数据集上优于最先进的方法。
英文原文摘要 1497 字符
Occluded visible-infrared person re-identification (Occluded VI-ReID) remains difficult due to modality heterogeneity and occlusions, both of which break structural consistency and weaken cross-modality feature alignment. Existing methods rely mainly on spatial-domain cues (such as local body parts and salient patches), but their discriminability degrades severely under varying imaging conditions or partial visibility. To address these issues, we introduce a spatial-frequency collaborative perspective that offers global perception and cross-location consistency. Specifically, we propose a Spatial-Frequency Collaborative Learning (SFCL) framework that uses frequency information to complement spatial representations. SFCL comprises a Cross-Modality Frequency Alignment Module (CFAM), a Spatial-Frequency Interaction Module (SFIM), and a Frequency-Aware Discriminative (FAD) loss. The CFAM models the spectral features of visible/infrared images in the frequency domain, establishing modality-consistent spectral priors. The SFIM injects these priors into spatial features, promoting dual-domain interaction and complementary representations of spatial and frequency semantics. In addition, the FAD loss jointly enforces cross-modality frequency alignment and semantic consistency, thus enhancing robustness and discriminability under occlusions. For real-occlusion evaluation, we construct two occluded datasets, Occ-SYSU-MM01 and Occ-RegDB, on which SFCL outperforms the state-of-the-art.