Publication Type : Conference Proceedings
Publisher : Springer Nature Switzerland
Source : IFIP Advances in Information and Communication Technology
Url : https://doi.org/10.1007/978-3-031-98356-6_10
Campus : Bengaluru
School : School of Computing
Year : 2025
Abstract : Gait recognition leverages unique walking patterns as biometric identifiers, enabling identification based on individual-specific gait signatures. Utilizing Convolutional Neural Networks (CNNs) and Gait Energy Images (GEIs), a model is developed to classify individuals through silhouette-based gait analysis. Both Conv1D and Conv2D CNN architectures are explored, with Conv2D yielding higher accuracy due to its enhanced ability to capture spatial features from 2D representations. The approach combines image processing techniques with deep learning to effectively distinguish individuals by their gait, demonstrating promising results for biometric identification. This work illustrates the adaptability of CNNs to dynamically help the patients with various neurological and age related disorders where gait recognition plays a major role. Also it would be of great help in collecting biometric data and reinforces their potential in security and identification applications.
Cite this Research Publication : K. Afnaan, Poluru Reddy Jahanve, Prajna Aasritha Adabala, Siwani Karna, Tripty Singh, Khaled Hushme, Leveraging Convolutional Neural Networks for Gait Recognition and Individual Identification for Improved Neurological Care, IFIP Advances in Information and Communication Technology, Springer Nature Switzerland, 2025, https://doi.org/10.1007/978-3-031-98356-6_10