This is a page that curates AI-related papers published worldwide. All content here is summarized using Google Gemini and operated on a non-profit basis. Copyright for each paper belongs to the authors and their institutions; please make sure to credit the source when sharing.
When Person Re-Identification Meets Event Camera: A Benchmark Dataset and An Attribute-guided Re-Identification Framework
Created by
Haebom
Author
Xiao Wang, Qian Zhu, Shujuan Wu, Bo Jiang, Shiliang Zhang, Yaowei Wang, Yonghong Tian, Bin Luo
Outline
In this paper, we present EvReID, a large-scale RGB-event-based person ReID dataset to address the data shortage problem in event camera-based person re-identification (ReID) research. EvReID contains 118,988 image pairs of 1,200 pedestrians, collected under various seasons, scenes, and lighting conditions. We also evaluate 15 state-of-the-art ReID algorithms to lay the foundation for future research. Furthermore, we propose TriPro-ReID, a contrastive learning framework that leverages pedestrian attributes to improve performance by exploiting pedestrian attributes as intermediate semantic features in addition to visual features from RGB frames and event streams. We verify the effectiveness of the proposed RGB-event-based person ReID framework through experiments on EvReID and MARS datasets. The dataset and source code will be made available at https://github.com/Event-AHU/Neuromorphic_ReID .
