PAC-GAN: An Effective Pose Augmentation Scheme for Unsupervised Cross-View Person Re-identification

Abstract

Person re-identification (person Re-Id) aims to retrieve the pedestrian images of a same person that captured by disjoint and non-overlapping cameras. Lots of researchers recently focuse on this hot issue and propose deep learning based methods to enhance the recognition rate in a supervised or unsupervised manner. However, two limitations that cannot be ignored: firstly, compared with other image retrieval benchmarks, the size of existing person Re-Id datasets are far from meeting the requirement, which cannot provide sufficient pedestrian samples for the training of deep model; secondly, the samples in existing datasets do not have sufficient human motions or postures coverage to provide more priori knowledges for learning. In this paper, we introduce a novel unsupervised pose augmentation cross-view person Re-Id scheme called PAC-GAN to overcome these limitations. We firstly present the formal definition of cross-view pose augmentation and then propose the framework of PAC-GAN that is a novel conditional generative adversarial network (CGAN) based approach to improve the performance of unsupervised corss-view person Re-Id. Specifically, The pose generation model in PAC-GAN called CPG-Net is to generate enough quantity of pose-rich samples from original image and skeleton samples. The pose augmentation dataset is produced by combining the synthesized pose-rich samples with the original samples, which is fed into the corss-view person Re-Id model named Cross-GAN. Besides, we use weight-sharing strategy in the CPG-Net to improve the quality of new generated samples. To the best of our knowledge, we are the first try to enhance the unsupervised cross-view person Re-Id by pose augmentation, and the results of extensive experiments show that the proposed scheme can combat the state-of-the-arts.

32 pages

References in corpus (7)

• Deep Learning in Neural Networks: An Overview
• Conditional Generative Adversarial Nets
• Joint Training of a Convolutional Network and a Graphical Model for Human Pose Estimation
• Multi-View Spectral Clustering via Structured Low-Rank Matrix Factorization
• Deep Adaptive Feature Embedding with Local Sample Distributions for Person Re-identification
• What-and-Where to Match: Deep Spatially Multiplicative Integration Networks for Person Re-identification
• Learning to Generate Images of Outdoor Scenes from Attributes and Semantic Layouts