Principal Investigator
MUHAMMAD WAQAR, MOHIBULLAH KHAN
Lecturer
(FE&CS)
Sign language is a visual language that uses hand gestures, facial expressions, and body movements to communicate. It is mostly used by deaf and hard-of-hearing individuals to communicate with each other and with those who understand the language. However, on contrary to its importance, developing a sign language for a particular local language becomes a very difficult task due to lack of standardization and complexity of language. For these reasons, there has been limited work on local Pakistani languages regarding their sign language recognition and translation. The proposed research work aims to create a deep learning framework that can be used to develop sign language recognition systems for specific local languages. The fine-tuned deep learning model will be trained on a custom dataset that can learn the underlying patterns and structures in sign language, and will effectively translate them into written or spoken language. The custom dataset can be used for future research in the given domain.
Completed Activites
1- Project Initiation:
Description: The project initiation phase involved defining objectives, assembling the research team, and outlining the scope and
methodology.
Ongoing Activites
2- Data Collection and Analysis:
Description: Development of a robust sign language recognition model trained on the custom dataset, achieving accurate detection of
signs.
3- AI Model Development:
Description: Development of a robust sign language recognition model trained on the custom dataset, achieving accurate detection of
signs.
Future Activites
4- Pilot Testing:
Description: Conducting user testing sessions to gather feedback on the effectiveness and usability of the recognition and translation
systems.
5- Development of Software Application
Description: Development of software application detect and translate signs
6- Paper Writeup
Description: Write up of research article highlighting major contribution of research work that may be submitted to well reputed journal.
Completetion Status
20% Completed