Real-time American Sign language recognition system | |
| Author | Lamichhane, Sandhya |
| Call Number | AIT Thesis no.DSAI-24-12 |
| Subject(s) | American Sign Language--Data processing Pattern recognition systems Natural language processing (Computer science) |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Data Science and Artificial Intelligence |
| Publisher | Asian Institute of Technology |
| Abstract | This paper develops a real-time American Sign Language (ASL) recognition system using ResNet-34andMViTv2-Sarchitectures. The models were trained on the WLASL100 dataset to classify 100 ASL signs, withMViTv2-S achieving a Top-1accuracy of 65.12% and faster inference times (168.2 ms) compared to ResNet-34 (178.33 ms). The system demonstrates potential for real-time applications due to its efficiency and accuracy. Challenges remain in scaling vocabulary size, addressing sentence-level recognition, and improving robustness to diverse conditions. This work provides a foundation for accessible communication tools for Deaf and Hard-of-Hearing communities. All codes and pre-trained model will be made available on github: https://github.com/creatorof/Real Time-Sign-Language-Recognition |
| Year | 2024 |
| Type | Thesis |
| School | School of Engineering and Technology |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Data Science and Artificial Intelligence (DSAI) |
| Chairperson(s) | Chaklam Silpasuwanchai; |
| Examination Committee(s) | Chantri Polprasert;Attaphongse Taparugssanagorn; |
| Scholarship Donor(s) | ADB-JSP; |
| Degree | Thesis (M. Sc.) - Asian Institute of Technology, 2024 |