Hard-Aware Point-to-Set Deep Metric for Person Re-identification

TL;DR

This paper introduces HAP2S, a novel deep metric learning loss function with hard-mining that improves person re-identification accuracy, robustness, and generality across multiple datasets and benchmarks.

Contribution

The paper proposes the HAP2S loss with a soft hard-mining scheme, enhancing deep metric learning for re-ID and other tasks by adaptively weighting harder samples.

Findings

HAP2S outperforms existing loss functions on re-ID benchmarks.

HAP2S demonstrates robustness to outliers.

HAP2S achieves state-of-the-art results on multiple datasets.

Abstract

Person re-identification (re-ID) is a highly challenging task due to large variations of pose, viewpoint, illumination, and occlusion. Deep metric learning provides a satisfactory solution to person re-ID by training a deep network under supervision of metric loss, e.g., triplet loss. However, the performance of deep metric learning is greatly limited by traditional sampling methods. To solve this problem, we propose a Hard-Aware Point-to-Set (HAP2S) loss with a soft hard-mining scheme. Based on the point-to-set triplet loss framework, the HAP2S loss adaptively assigns greater weights to harder samples. Several advantageous properties are observed when compared with other state-of-the-art loss functions: 1) Accuracy: HAP2S loss consistently achieves higher re-ID accuracies than other alternatives on three large-scale benchmark datasets; 2) Robustness: HAP2S loss is more robust to outliers than other losses; 3) Flexibility: HAP2S loss does not rely on a specific weight function, i.e., different instantiations of HAP2S loss are equally effective. 4) Generality: In addition to person re-ID, we apply the proposed method to generic deep metric learning benchmarks including CUB-200-2011 and Cars196, and also achieve state-of-the-art results.