Search papers, labs, and topics across Lattice.
This paper introduces MDCRNet, a Multi-scale Decomposed Convolution Refinement Network designed to address the challenges of visible-infrared person re-identification (VI-ReID) by enhancing cross-modal feature learning and discriminative metric learning. The architecture employs a Hierarchical Learning Module with four Hierarchical Decomposed Convolution Attention modules, which utilize lightweight channel attention and multi-scale spatial perception to effectively capture spatial dependencies. Experimental results on the SYSU-MM01 and RegDB datasets show that MDCRNet achieves state-of-the-art performance, significantly improving recognition accuracy across modalities.
MDCRNet shatters previous performance barriers in visible-infrared person re-identification by leveraging multi-scale spatial dependencies and a novel discriminative loss framework.
Visible-infrared person re-identification (VI-ReID) suffers from cross-modal discrepancies and limited discriminative capabilities, leading to suboptimal recognition performance. Current approaches exhibit limitations in semantic mining, cross-modal fusion and feature constraints. To tackle these challenges, we propose MDCRNet, a Multi-scale Decomposed Convolution Refinement Network that enhances cross-modal feature learning and discriminative metric learning. Specifically, we introduce a Hierarchical Learning Module (HLM) containing four Hierarchical Decomposed Convolution Attention (HDCA) modules, each equipped with lightweight channel attention and multi-scale spatial perception blocks to capture multi-scale spatial dependencies. Moreover, we develop a Joint Discriminative Metric Loss (JDML) incorporating a novel Granularity Discriminative Loss (GDL) that simultaneously optimizes intra-identity compactness and inter-identity separability across modalities. Extensive experiments on SYSU-MM01 and RegDB datasets demonstrate that MDCRNet achieves state-of-the-art performance on both benchmarks. Code is available at https://github.com/Kevin-zms/MDCRNet.
