Neural Image Parts Group Search for Person Re-Identification

Employing partition strategy to explore fine-grained features has been verified to be beneficial for person re-identification in recent literature. However, existing methods primarily rely on expert experience to manually design various partition strategies, which may lead to a sub-optimal solution for fine-grained features exploration. In this paper, we propose a Neural Parts Group Search (NPGS) strategy that auto-searches the optimal parts group via evolutionary algorithm (EA) to facilitate the network to exploit the local details. And during search process, designing a high-quality search space is especially crucial for an efficient optimization. Considering the human top-down structure and the semantic coherence of parts, we design a coarse-to-fine parts search space (C2F-PSP) in NPGS, which effectively reduce the search complexity without the loss of parts expressivity. Additionally, since only employing the high-level semantic features is insufficient for the NPGS to search effective parts, we further develop an efficient feature aggregation strategy named hierarchical low-rank bilinear pooling that progressively integrates the high-level semantic property and the low-level fine-grained details to facilitate the NPGS to explore the fine-grained features. Furthermore, to relieve the interference of background during parts search process, we propose a novel Relational Attention Module (RAM) by exploiting the channel and spatial structural interdependence of pixels to strengthen the discriminative regions. Extensive experiments on the mainstream evaluation datasets demonstrate that our method outperforms the recent state-of-the-art Re-ID models.

Automated summary

This paper proposes a Neural Parts Group Search (NPGS) strategy for person re-identification. The optimal parts group is auto-searched using an evolutionary algorithm, allowing the network to exploit local details. A coarse-to-fine parts search space (C2F-PSP) is designed to reduce search complexity without losing parts expressivity. An efficient feature aggregation strategy named hierarchical low-rank bilinear pooling is developed to integrate high-level semantic and low-level fine-grained details. A relational attention module (RAM) is proposed to reduce interference from the background during parts search. Experimental results show that the proposed method outperforms state-of-the-art Re-ID models.