Methods of Gait Recognition in Video


如何引用文章

全文:

开放存取 开放存取
受限制的访问 ##reader.subscriptionAccessGranted##
受限制的访问 订阅存取

详细

Human gait is an important biometric index that allows to identify a person at a great distance without direct contact. Due to these qualities, which other popular identifiers such as fingerprints or iris do not have, the recognition of a person by the manner of walking has become very common in various areas where video surveillance systems can be used. With the development of computer vision techniques, a variety of approaches for human identification by movements in a video appear. These approaches are based both on natural biometric characteristics (human skeleton, silhouette, and their change during walking) and abstract features trained automatically which do not have physical justification. Modern methods combine classical algorithms of video and image analysis and new approaches that show excellent results in related tasks of computer vision, such as human identification by face and appearance or action and gesture recognition. However, due to the large number of conditions that can affect the walking manner of a person itself and its representation in video, the problem of identifying a person by gait still does not have a sufficiently accurate solution. Many methods are overfitted by the conditions presented in the databases on which they are trained, which limits their applicability in real life. In this paper, we provide a survey of state-of-the-art methods of gait recognition, their analysis and comparison on several popular video collections and for different formulations of the problem of recognition. We additionally reveal the problems that prevent the final solution of gait identification challenge.

作者简介

A. Sokolova

National Research University Higher School of Economics

编辑信件的主要联系方式.
Email: ale4kasokolova@gmail.com
俄罗斯联邦, Moscow, 101000

A. Konushin

National Research University Higher School of Economics; Lomonosov Moscow State University

编辑信件的主要联系方式.
Email: anton.konushin@grapics.cs.msu.ru
俄罗斯联邦, Moscow, 101000; Moscow, 119991

补充文件

附件文件
动作
1. JATS XML

版权所有 © Pleiades Publishing, Ltd., 2019