Submit

Login Login

Register Register

Home / Journals / CMC / Online First / doi:10.32604/cmc.2025.075704
Journal Menu
Special Issues
Table of Content

All issues

Online First

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Open Access

ARTICLE

GaitMAFF: Adaptive Multi-Modal Fusion of Skeleton Maps and Silhouettes for Robust Gait Recognition in Complex Scenarios

Zhongbin Luo1,2, Zhaoyang Guan3, Wenxing You2, Yunteng Wang2, Yanqiu Bi4,5,*
1 College of Computer Science, Chongqing University, Chongqing, 400044, China
2 School of Traffic and Transportation, Chongqing Jiaotong University, Chongqing, 400074, China
3 Department of Engineering Sciences and Applied Mathematics, Northwestern University, Evanston, IL 60208, USA
4 School of Civil Engineering, Chongqing Jiaotong University, Chongqing, 400074, China
5 National & Local Joint Engineering Research Center of Transportation Civil Engineering Materials, Chongqing Jiaotong University, Chongqing, 400074, China
* Corresponding Author: Yanqiu Bi. Email: email

Computers, Materials & Continua https://doi.org/10.32604/cmc.2025.075704

Received 06 November 2025; Accepted 15 December 2025; Published online 30 December 2025

Abstract

Gait recognition is a key biometric for long-distance identification, yet its performance is severely degraded by real-world challenges such as varying clothing, carrying conditions, and changing viewpoints. While combining silhouette and skeleton data is a promising direction, effectively fusing these heterogeneous modalities and adaptively weighting their contributions in response to diverse conditions remains a central problem. This paper introduces GaitMAFF, a novel Multi-modal Adaptive Feature Fusion Network, to address this challenge. Our approach first transforms discrete skeleton joints into a dense Skeleton Map representation to align with silhouettes, then employs an attention-based module to dynamically learn the fusion weights between the two modalities. These fused features are processed by a powerful spatio-temporal backbone with Weighted Global-Local Feature Fusion Modules (WFFM) to learn a discriminative representation. Extensive experiments on the challenging CCPG and Gait3D datasets show that GaitMAFF achieves state-of-the-art performance, with an average Rank-1 accuracy of 84.6% on CCPG and 58.7% on Gait3D. These results demonstrate that our adaptive fusion strategy effectively integrates complementary multi-modal information, significantly enhancing gait recognition robustness and accuracy in complex scenes and providing a practical solution for real-world applications.

Keywords

Gait recognition; multi-modal fusion; adaptive feature fusion; skeleton map; silhouette

  • 298

    View

  • 46

    Download

  • 1

    Like

Related articles
Copyright© 2026 Tech Science Press
© 1997-2026 TSP (Henderson, USA) unless otherwise stated

Share Link