An Applied Holistic Landmark with Deep Learning for Thai Sign Language Recognition

Thai Sign Language is the main language for communication with the hearing impaired in Thailand. It uses hand or finger signs or gestures, as well as facial expressions and body movements, to convey the meaning of spoken words to people who are deaf or hard of hearing. This research is an attempt to develop a Thai Sign Language interpreter with the holistic approach of landmark detection for hand, facial, and body movements with various types of deep learning algorithms for both recognition and classification. With this approach, we propose a method to recognize the Thai sign language using holistic landmarks with deep learning algorithms. The result shows we can have an accuracy of over 90 percent.