Multi-scale Decomposed Convolution Refinement Network for Visible-Infrared Person Re-Identification
2026-08-17 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors address the problem of recognizing people across visible and infrared images, which is hard because the two types of images look very different and are hard to compare. They develop a new network called MDCRNet that improves how features from both image types are learned and distinguished. Their method uses special attention techniques to capture details at different scales and a new loss function to better separate people in the feature space. Tests on two common datasets show their approach works better than previous methods.
visible-infrared person re-identificationcross-modal discrepanciesmulti-scale spatial perceptionchannel attentionmetric learningdiscriminative lossfeature fusionSYSU-MM01 datasetRegDB dataset
Authors
Mingsheng Zheng, Zirui Jiang, Bo Liu, Yupeng Chen, Jun Zhang, Kai Zhao
Abstract
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.