Scalable Person Re-identification on Supervised Smoothed Manifold

TL;DR

This paper introduces a scalable manifold-based affinity learning algorithm for person re-identification that leverages supervision, enhances accuracy, and is efficient enough for large-scale real-world applications.

Contribution

The paper proposes a novel manifold-preserving algorithm that utilizes pairwise supervision, scales efficiently, and can improve existing re-identification methods as a postprocessing step.

Findings

Outperforms state-of-the-art on CUHK03 and Market-1501 datasets.

Effectively utilizes pairwise supervision for better manifold modeling.

Achieves high efficiency suitable for large-scale applications.

Abstract

Most existing person re-identification algorithms either extract robust visual features or learn discriminative metrics for person images. However, the underlying manifold which those images reside on is rarely investigated. That raises a problem that the learned metric is not smooth with respect to the local geometry structure of the data manifold. In this paper, we study person re-identification with manifold-based affinity learning, which did not receive enough attention from this area. An unconventional manifold-preserving algorithm is proposed, which can 1) make the best use of supervision from training data, whose label information is given as pairwise constraints; 2) scale up to large repositories with low on-line time complexity; and 3) be plunged into most existing algorithms, serving as a generic postprocessing procedure to further boost the identification accuracies. Extensive experimental results on five popular person re-identification benchmarks consistently demonstrate the effectiveness of our method. Especially, on the largest CUHK03 and Market-1501, our method outperforms the state-of-the-art alternatives by a large margin with high efficiency, which is more appropriate for practical applications.