WildGait: Learning Gait Representations from Raw Surveillance Streams

Abstract

The use of gait for person identification has important advantages such as being non-invasive, unobtrusive, not requiring cooperation and being less likely to be obscured compared to other biometrics. Existing methods for gait recognition require cooperative gait scenarios, in which a single person is walking multiple times in a straight line in front of a camera. We aim to address the challenges of real-world scenarios in which camera feeds capture multiple people, who in most cases pass in front of the camera only once. We address privacy concerns by using only motion information of walking individuals, with no identifiable appearance-based information. As such, we propose a novel weakly supervised learning framework, WildGait, which consists of training a Spatio-Temporal Graph Convolutional Network on a large number of automatically annotated skeleton sequences obtained from raw, real-world, surveillance streams to learn useful gait signatures. We collected the training data and compiled the largest dataset of walking skeletons called Uncooperative Wild Gait, containing over 38k tracklets of anonymized walking 2D skeletons. We release the dataset for public use. Our results show that, with fine-tuning, we surpass the current state-of-the-art pose-based gait recognition solutions. Our proposed method is reliable in training gait recognition methods in unconstrained environments, especially in settings with scarce amounts of annotated data.

arXiv comment9 pages, 7 figures, 5 Tables

References in corpus (14)

• FaceNet: A Unified Embedding for Face Recognition and Clustering
• Overcoming catastrophic forgetting in neural networks
• A Simple Framework for Contrastive Learning of Visual Representations
• Simple Online and Realtime Tracking
• The Kinetics Human Action Video Dataset
• OpenPose: Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields
• Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition
• Monocular Human Pose Estimation: A Survey of Deep Learning-based Methods
• MS${}^2$L: Multi-Task Self-Supervised Learning for Skeleton Based Action Recognition
• Supervised Contrastive Learning
• Self-supervised Video Representation Learning by Pace Prediction
• CrowdPose: Efficient Crowded Scenes Pose Estimation and A New Benchmark
• Iterate & Cluster: Iterative Semi-Supervised Action Recognition
• PREDICT & CLUSTER: Unsupervised Skeleton Based Action Recognition