基于特定和共享表示学习的可见光与近红外行人重识别方法

By using a dual-branch shared specific joint learning network, the modal differences in visible and near-infrared images are mitigated. The ResNet-50 module is used to extract specific and shared features, and the feature distribution is optimized through intra-class aggregation and inter-class separation learning. This solves the problem of inconsistent modal features in cross-modal person re-identification, and improves recognition accuracy and generalization performance.

CN117523609BActive Publication Date: 2026-07-17ANHUI UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
ANHUI UNIV
Filing Date
2023-11-15
Publication Date
2026-07-17

Smart Images

  • Figure CN117523609B_ABST
    Figure CN117523609B_ABST
Patent Text Reader

Abstract

本发明公开一种基于特定和共享表示学习的可见光与近红外行人重识别方法，设计一个共享分支来弥合图像级别的域差异并学习模态共享表征，同时设计了一个特定分支来保留可见光图像的判别信息以学习模态特定表征。此外，还提出了类内聚合和类间分离学习策略，以在细粒度水平上优化特征嵌入的分布，进一步提高发明方法的泛化性能，本发明能够在消除颜色差异的同时，保留可见光图像中包含的颜色特定信息。
Need to check novelty before this filing date? Find Prior Art