Origins method improves gait recognition with unified template learning
Learning A Unified Template for Gait Recognition
Computer Vision and Pattern Recognition
Summary
Gait recognition means identifying people by the way they walk, which is tricky because walking changes with different conditions. The authors created a new method called Origins that learns a single, unified template to represent all walking styles and conditions. By using ideas from image generation called Diffusion Models, Origins can both generate and recognize walking patterns more consistently. This approach helps computers tell people apart more accurately by focusing on key walking differences within a stable framework.
What this means in practice
- •For security system engineers: Build more reliable gait recognition modules that accurately identify individuals under various conditions by using a unified template approach.
- •For surveillance technology developers: Enhance person identification in video analytics by integrating generative and representation learning to handle diverse walking patterns.
Authors
Panjian Huang, Saihui Hou, Junzhou Huang, Yongzhen Huang
Abstract
"What I cannot create, I do not understand."Human wisdom reveals that creation is one of the highest forms of learning. For example, Diffusion Models have demonstrated remarkable semantic structure and memory in image generation, understanding, and restoration, which intuitively benefits representation learning. However, current gait networks rarely embrace this perspective, relying primarily on learning by contrasting gait samples under varying complex conditions, leading to semantic inconsistency and uniformity issues. To address these issues, we propose Origins with generative capabilities whose underlying philosophy is that different entities are generated from a unified template, inherently regularizing gait representations within a consistent and diverse semantic space to capture accurate gait differences. Admittedly, learning this unified template is exceedingly challenging, as it requires the comprehensiveness of the template to encompass gait representations with various conditions. Inspired by Diffusion Models, Origins diffuses the unified template into timestep templates for gait generative learning, and meanwhile transfers the unified template for gait representation learning. Especially, gait generative and representation learning serve as a unified framework for end-to-end joint training. Extensive experiments on CASIA-B, CCPG,SUSTech1K, Gait3D, GREW and CCGR-MINI demonstrate that Origins performs unified generative and representation learning, achieving superior performance.