Human Identification Based on Gait Representation and Analysis
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Human identification based on gait analysis is a promising biometric technology that can recognize differentindividuals by their walking patterns. This study primarily addresses the challenges of gait representation andpartial occlusion. Firstly, considering the multi-scale and multi-perspective aspects of gait in practical applicationscenarios, a novel block collaborative gait representation method is proposed based upon local structures,aiming to enhance the accuracy of identity recognition by integrating information from multiple scales andperspectives. Then, we propose a new gait recognition network that integrates dilated convolutions and theresidual mechanism (DCRM). The DCRM network adds dilated convolutional blocks to the residual branch toexpand the receptive field without losing resolution, thereby reducing the negative impact of local occlusionon recognition accuracy. Experimental results on two public datasets demonstrated that the proposed approachshows clear advantages over existing gait analysis methods.
Abstract
Human identification based on gait analysis is a promising biometric technology that can recognize differentindividuals by their walking patterns. This study primarily addresses the challenges of gait representation andpartial occlusion. Firstly, considering the multi-scale and multi-perspective aspects of gait in practical applicationscenarios, a novel block collaborative gait representation method is proposed based upon local structures,aiming to enhance the accuracy of identity recognition by integrating information from multiple scales andperspectives. Then, we propose a new gait recognition network that integrates dilated convolutions and theresidual mechanism (DCRM). The DCRM network adds dilated convolutional blocks to the residual branch toexpand the receptive field without losing resolution, thereby reducing the negative impact of local occlusionon recognition accuracy. Experimental results on two public datasets demonstrated that the proposed approachshows clear advantages over existing gait analysis methods.
참고문헌 (16)
- A. Sepas-Moghaddam, "Deep gait recognition : a survey", IEEE Transactions on Pattern Analysis and Machine Intelligence, 45(1), pp.264~284 (2023)[DOI]
- J. N. Mogan, "Advances in vision-based gait recognition : from handcrafted to deep learning", Sensors, 22(15) (2022)[DOI]
- X. Wang, "Human identification based on gait manifold", Applied Intelligence, 53(5), pp.6062~6073 (2023)[DOI]
- J. Amin, "Convolutional Bi-LSTM based human gait recognition using video sequences", Computers, Materials & Continua, 68(2), pp.2693~2709 (2021)[DOI]
- T. Chai, "Lagrange motion analysis and view embeddings for improved gait recognition", Proceedings of the IEEE, 2022, 20217~20226
- M. A. A. Faisal, "NDDNet : a deep learning model for predicting neurodegenerative diseases from gait pattern", Applied Intelligence, 53(17), pp.20034~20046 (2023)[DOI]
- X. Chen, "Multi-view gait image generation for cross-view gait recognition", IEEE Transactions on Image Processing, 30, pp.3041~3055 (2021)[DOI]
- X. Wang, "Visual gait recognition based on convolutional block attention network", Multimedia Tools and Applications, 81(20), pp.29459~29476 (2022)[DOI]
- J. Wen, "Cross-view gait recognition based on residual long short-term memory", Multimedia Tools and Applications, 80(19), pp.28777~28788 (2021)[DOI]
- Y. Wang, "Posture estimation & posture sequence recognition for martial artists", Human-centric Computing and Information Sciences, 13 (2023)
- J. Wen, "Gait recognition based on GF-CNN and metric learning", Journal of Information Processing Systems, 16(5), pp.1105~1112 (2020)[DOI]
- H. Chao, "GaitSet : cross-view gait recognition through utilizing gait as a deep set", IEEE Transactions on Pattern Analysis and Machine Intelligence, 44(7), pp.3467~3478 (2022)[DOI]
- X. Wang, "Gait classification through CNN-based ensemble learning", Multimedia Tools and Applications, 80(1), pp.1565~1581 (2021)[DOI]
- J. Wan, "Omni-domain feature extraction method for gait recognition", Mathematics, 11(12) (2023)[DOI]
- A. Mehmood, "Human gait recognition by using two stream neural network along with spatial and temporal features", Pattern Recognition Letters, 180, pp.16~25 (2024)[DOI]
- X. Huang, "Condition-adaptive graph convolution learning for skeleton-based gait recognition", IEEE Transactions on Image Processing, 32, pp.4773~4784 (2023)[DOI]