WITHDRAWN: Hierarchical Attentive Feature Aggregation for Person Re-identification

Abstract

Recent efforts on person re-identification have shown promising results by learning discriminative features via multi-branch network. To further boost feature discrimination, attention mechanism has also been extensively employed. However, the branches in existing models generally work independently, which may compromise the ability of mining diverse features. To mitigate this issue, a novel framework called Hierarchical Attentive Feature Aggregation (Hi-AFA) is proposed. In Hi-AFA, a hierarchical aggregation mechanism is applied to learn attentive features. The current feature map is not only fed into the next stage, but also aggerated into another branch, leading to hierarchical feature flows along depth and parallel branches. We also present a simple feature suppression operation and a lightweight dual attention module with only a few parameters to guide feature learning. By this manner, the branches could cooperate to mine richer and more diverse feature representations. We integrate the hierarchical aggregation and multi-granularity feature learning into a unified architecture that builds upon OSNet, resulting a resource-economical and effective person re-identification model. Extensive experiments on three public datasets, including Market-1501, DukeMTMC-reID, and CUHK03, are conducted to validate the effectiveness of the proposed method, and results show that state-of-the-art performance is achieved.