GaitRGA: Gait Recognition Based on Relation-Aware Global Attention

Gait recognition, a long-range biometric technique based on walking posture, the fact that they do not require the cooperation of the subject and are non-invasive has made them highly sought after in recent years.Although existing methods have achieved impressive results in laboratory environments,...

Full description

Saved in:
Bibliographic Details
Main Authors: Jinhang Liu, Yunfan Ke, Ting Zhou, Yan Qiu, Chunzhi Wang
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/8/2337
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850154756795793408
author Jinhang Liu
Yunfan Ke
Ting Zhou
Yan Qiu
Chunzhi Wang
author_facet Jinhang Liu
Yunfan Ke
Ting Zhou
Yan Qiu
Chunzhi Wang
author_sort Jinhang Liu
collection DOAJ
description Gait recognition, a long-range biometric technique based on walking posture, the fact that they do not require the cooperation of the subject and are non-invasive has made them highly sought after in recent years.Although existing methods have achieved impressive results in laboratory environments, the recognition performance is still deficient in real-world applications, especially when confronted with complex and dynamic scenarios. The major challenges in gait recognition include changes in viewing angle, occlusion, clothing changes, and significant differences in gait characteristics under different walking conditions. To slove these issues, we propose a gait recognition method based on relational-aware global attention. Specifically, we introduce a Relational-aware Global Attention (RGA) module, which captures global structural information within gait sequences to enable more precise attention learning. Unlike traditional gait recognition methods that rely solely on local convolutions, we stack pairwise associations between each feature position in the gait silhouette and all other feature positions, along with the features themselves, using a shallow convolutional model to learn attention. This approach is particularly effective in gait recognition due to the physical constraints on human walking postures, allowing the structural information embedded in the global relationships to aid in inferring the semantics and focus areas of various body parts, thereby improving the differentiation of gait features across individuals. Our experimental results on multiple datasets (Grew, Gait3D, SUSTech1k) demonstrate that GaitRGA achieves significant performance improvements, especially in real-world scenarios.
format Article
id doaj-art-8d13b5180a354801879b8bf9b5e6a42c
institution OA Journals
issn 1424-8220
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-8d13b5180a354801879b8bf9b5e6a42c2025-08-20T02:25:12ZengMDPI AGSensors1424-82202025-04-01258233710.3390/s25082337GaitRGA: Gait Recognition Based on Relation-Aware Global AttentionJinhang Liu0Yunfan Ke1Ting Zhou2Yan Qiu3Chunzhi Wang4School of Computer Science, Hubei University of Technology, Wuhan 430068, ChinaSchool of Computer Science, Hubei University of Technology, Wuhan 430068, ChinaSchool of Computer Science, Hubei University of Technology, Wuhan 430068, ChinaDepartment of Computer Science, College of Engineering and Technology, Hubei University of Technology, Wuhan 430068, ChinaSchool of Computer Science, Hubei University of Technology, Wuhan 430068, ChinaGait recognition, a long-range biometric technique based on walking posture, the fact that they do not require the cooperation of the subject and are non-invasive has made them highly sought after in recent years.Although existing methods have achieved impressive results in laboratory environments, the recognition performance is still deficient in real-world applications, especially when confronted with complex and dynamic scenarios. The major challenges in gait recognition include changes in viewing angle, occlusion, clothing changes, and significant differences in gait characteristics under different walking conditions. To slove these issues, we propose a gait recognition method based on relational-aware global attention. Specifically, we introduce a Relational-aware Global Attention (RGA) module, which captures global structural information within gait sequences to enable more precise attention learning. Unlike traditional gait recognition methods that rely solely on local convolutions, we stack pairwise associations between each feature position in the gait silhouette and all other feature positions, along with the features themselves, using a shallow convolutional model to learn attention. This approach is particularly effective in gait recognition due to the physical constraints on human walking postures, allowing the structural information embedded in the global relationships to aid in inferring the semantics and focus areas of various body parts, thereby improving the differentiation of gait features across individuals. Our experimental results on multiple datasets (Grew, Gait3D, SUSTech1k) demonstrate that GaitRGA achieves significant performance improvements, especially in real-world scenarios.https://www.mdpi.com/1424-8220/25/8/2337GaitRGAbiometric identificationsilhouette-based Gait recognitiondeep learningneural network
spellingShingle Jinhang Liu
Yunfan Ke
Ting Zhou
Yan Qiu
Chunzhi Wang
GaitRGA: Gait Recognition Based on Relation-Aware Global Attention
Sensors
GaitRGA
biometric identification
silhouette-based Gait recognition
deep learning
neural network
title GaitRGA: Gait Recognition Based on Relation-Aware Global Attention
title_full GaitRGA: Gait Recognition Based on Relation-Aware Global Attention
title_fullStr GaitRGA: Gait Recognition Based on Relation-Aware Global Attention
title_full_unstemmed GaitRGA: Gait Recognition Based on Relation-Aware Global Attention
title_short GaitRGA: Gait Recognition Based on Relation-Aware Global Attention
title_sort gaitrga gait recognition based on relation aware global attention
topic GaitRGA
biometric identification
silhouette-based Gait recognition
deep learning
neural network
url https://www.mdpi.com/1424-8220/25/8/2337
work_keys_str_mv AT jinhangliu gaitrgagaitrecognitionbasedonrelationawareglobalattention
AT yunfanke gaitrgagaitrecognitionbasedonrelationawareglobalattention
AT tingzhou gaitrgagaitrecognitionbasedonrelationawareglobalattention
AT yanqiu gaitrgagaitrecognitionbasedonrelationawareglobalattention
AT chunzhiwang gaitrgagaitrecognitionbasedonrelationawareglobalattention