Study on Hand Keypoint Framework for Sign Language Recognition
Sholikhatul Amaliya, Anik Nur Handayani, Muhammad Akbar, Heru Wahyu Herwanto, Osamu Fukuda, Wendy Cahya Kurniawan · 2021 7th International Conference on Electrical, Electronics and Information Engineering (ICEEIE) · 2021
This study looks at the development of a sign language translation system that will serve as a communication link between deaf and hearing people. The Indonesian Language Sign System, called SIBI, is one of the communication techniques used by people with hearing impairment in Indonesia. Communication between deaf persons and non-deaf people has been found to have several issues, including a lack of community awareness of sign language. With this background, the presence of hand keypoint detection is one solution that can be implemented. OpenPose, TensorfowLite, and MediaPipe are just a few of the frameworks that are frequently used to identify hand keypoints. The goal of this research is to compare the best framework for working with the picture dataset generated by the Kinect 2.0 camera. The findings of this study show that the TensorflowLite framework can generate tidy hand keypoints and can be applied effectively to the dataset obtained.