Logo image
Back
HAMoBE: Hierarchical and Adaptive Mixture of Biometric Experts for Video-Based Person ReID
Conference proceeding

HAMoBE: Hierarchical and Adaptive Mixture of Biometric Experts for Video-Based Person ReID

Yiyang Su, Yunping Shi, Feng Liu and Xiaoming Liu
Proceedings / IEEE International Conference on Computer Vision, pp 11525-11536
19 Oct 2025
DOI: https://doi.org/10.1109/ICCV51701.2025.01072

Abstract

Antennas Broadcast technology Broadcasting Feeds Filtering Filters hierarchical and adaptive mixture of biometric experts (hamobe) Millimeter wave integrated circuits MIMICs Monolithic integrated circuits Protocols video-based person reid
Recently, research interest in person re-identification (ReID) has increasingly focused on video-based scenarios, essential for robust surveillance and security in varied and dynamic environments. However, existing videobased ReID methods often overlook the necessity of identifying and selecting the most discriminative features from both videos in a query-gallery pair for effective matching. To address this issue, we propose a novel Hierarchical and Adaptive Mixture of Biometric Experts (HAMoBE) framework, which leverages multi-layer features from a pretrained large model (e.g., CLIP) and is designed to mimic human perceptual mechanisms by independently modeling key biometric features-appearance, static body shape, and dynamic gait-and adaptively integrating them. Specifically, HAMoBE includes two levels: the first level extracts low-level features from multi-layer representations provided by the frozen large model, while the second level consists of specialized experts focusing on long-term, shortterm, and temporal features. To ensure robust matching, we introduce a new dual-input decision gating network that dynamically adjusts the contributions of each expert based on their relevance to the input scenarios. Extensive evaluations on benchmarks like MEVID demonstrate that our approach yields significant performance improvements (e.g., +13.0 % Rank1).

Metrics

1 Record Views

Details

Logo image