{"id":20177405,"url":"https://github.com/saba-gul/gait-analysis-using-3d-cnn","last_synced_at":"2026-06-06T04:31:47.887Z","repository":{"id":221823918,"uuid":"755483632","full_name":"Saba-Gul/Gait-Analysis-Using-3D-CNN","owner":"Saba-Gul","description":"Gait Recognition with 3D CNN. This project proposes a novel approach using 3D convolutional neural networks (3D CNN) to capture spatio-temporal features of gait sequences for robust recognition in an un-intrusive manner.","archived":false,"fork":false,"pushed_at":"2024-03-11T14:18:25.000Z","size":1906,"stargazers_count":2,"open_issues_count":0,"forks_count":1,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-01-13T16:28:21.714Z","etag":null,"topics":["3d-convolutional-network","baysian-optimisation","cnn-classification","computer-vision","gait","gait-analysis","gait-energy-image","gait-recognition","optimization-algorithms"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Saba-Gul.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-02-10T10:56:34.000Z","updated_at":"2024-11-20T18:37:02.000Z","dependencies_parsed_at":null,"dependency_job_id":"66b10ff2-8c5a-4a98-abb7-bbf4f1db0cba","html_url":"https://github.com/Saba-Gul/Gait-Analysis-Using-3D-CNN","commit_stats":null,"previous_names":["saba-gul/gait-analysis-using-3d-cnn"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Saba-Gul%2FGait-Analysis-Using-3D-CNN","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Saba-Gul%2FGait-Analysis-Using-3D-CNN/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Saba-Gul%2FGait-Analysis-Using-3D-CNN/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Saba-Gul%2FGait-Analysis-Using-3D-CNN/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Saba-Gul","download_url":"https://codeload.github.com/Saba-Gul/Gait-Analysis-Using-3D-CNN/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":241612340,"owners_count":19990774,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["3d-convolutional-network","baysian-optimisation","cnn-classification","computer-vision","gait","gait-analysis","gait-energy-image","gait-recognition","optimization-algorithms"],"created_at":"2024-11-14T02:15:41.973Z","updated_at":"2025-03-03T05:22:53.673Z","avatar_url":"https://github.com/Saba-Gul.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Gait Recognition with 3D CNN\n\n## Introduction\nGait recognition is a promising avenue for identification and authentication due to the uniqueness of individual strides. This project proposes a novel approach using 3D convolutional neural networks (3D CNN) to capture spatio-temporal features of gait sequences for robust recognition.\n\n## Methodology\n- **3D CNN Architecture**: The proposed network architecture employs a holistic approach using gait energy images (GEI) to capture shape and motion (Spatio-Temporal features) characteristics of human gait.\n \n ![Gait Analysis1](images/Proposed_framework.png \"3D Convolutional Neural Network for Inter-class Subject Identification\")\n\n- **Dataset**: Evaluation was conducted on two publicly available datasets, OULP and CASIA-B, which exhibit substantial gender and age diversity.\n  \n ![Gait Analysis2](images/CASIA-B_Dataset.png \"Data-set Specifications CASIA-B\")\n\n ![Gait Analysis3](images/OULP_Dataset.png \"Data-set Specifications OULP\")\n\n- **Optimization Strategies**: Bayesian algorithms were explored to tune hyperparameters and enhance network performance.\n\n ![Gait Analysis4](images/HypOpt.png \"Hyper-parameter tuning using bayesian optimization\")\n\n## Key Features\n- Robust Gait Recognition: The optimized 3D CNN and GEI effectively capture unique gait characteristics despite challenging covariates such as change in speed, viewpoint, clothing, and carrying accessories.\n- State-of-the-Art Results: Achieved state-of-the-art results on multi-views and multiple carrying conditions of subjects in the CASIA-B dataset.\n  \n ![Gait Analysis7](images/Performance.png \"Comparison of our framework with state of art models\")\n \n ![Gait Analysis5](images/CASIAB_Results.png \"Proposed network training results on CASIA-B dataset\")\n\n ![Gait Analysis6](images/OULP_Results.png \"Proposed network training results on OULP dataset\")\n \n\n \n## Future Directions\n- Overcoming Overfitting: Address potential overfitting issues due to limited variance and frames per subject in the OULP dataset.\n- Genetic Optimization Algorithms: Explore genetic optimization algorithms to further enhance performance.\n- Real-life Scenarios: Extend the framework to practical environments by tackling more challenging real-life scenarios for person identification based on walking patterns.\n\n## Usage\n- Clone the repository.\n- Install necessary dependencies.\n- Execute the main script for gait recognition.\n\n## Citation\nGul S., Malik M.I., Khan G.M., Shafait F. (2021) Multi-view Gait Recognition System using\nSpatio-temporal Features and Deep learning, Expert Systems with Applications,115057, ISSN\n0957-4174, https://doi.org/10.1016/j.eswa.2021.115057.\n\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsaba-gul%2Fgait-analysis-using-3d-cnn","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsaba-gul%2Fgait-analysis-using-3d-cnn","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsaba-gul%2Fgait-analysis-using-3d-cnn/lists"}