| Proj No. | A2047-251 |
| Title | Data and Model Architecture Modality Exploration for Gait Detection and Recognition |
| Summary | Gait is an essential biometric that has the unique advantage of human identification at a distance without physical contact. Gait recognition is a promising biometric technique for human identification in surveillance and security applications. Compared to traditional appearance-centric approaches, recent advances in learning-based methods offer opportunities to improve robustness under challenging conditions such as low light, occlusion, viewpoint variation, clothing changes, and long-distance scenarios. However, many existing solutions process data frame-by-frame or rely on narrowly scoped pipelines, overlooking important factors such as sequence quality, motion consistency, and natural variability in walking patterns. This project aims to explore and develop general representations and models for gait detection and recognition, with an emphasis on practical robustness and adaptability rather than a single fixed technique. In this project, the student is required to investigate and implement advanced deep learning approaches to extract both motion and biometric cues for gait analysis. The student could start by surveying representative architectures and pipeline designs, establishing baselines, and progressively refining model components, training strategies, and evaluation protocols. The developed methods will be evaluated on publicly available gait datasets using standard metrics and clearly defined benchmarks. The students who wish to work on this project are strongly encouraged to contact the supervisor for more information before making the selection. |
| Supervisor | Prof Chang Chip Hong (Loc:S2 > S2 B2C > S2 B2C 97, Ext: +65 67905873) |
| Co-Supervisor | - |
| RI Co-Supervisor | - |
| Lab | Centre for Integrated Circuits & Systems (CICS) (Loc: S3.2-B2-05) |
| Single/Group: | Single |
| Area: | Smart Electronics and IC design |
| ISP/RI/SMP/SCP?: |