What is it about?
We developed, implemented, and evaluated a promising strategy to automatically recognize Brazilian Sign Language signs from video sequences, using a combination of Digital Image Processing techniques and Artificial Intelligence. The proposed system performs the recognition of the signs using hand configuration, orientation, point of articulation and movement as parameters.
Featured Image
Photo by Thiago Barletta on Unsplash
Why is it important?
The deaf community faces a major obstacle in their daily routine: the difficulty in communicating with the predominantly hearing society. In recent years, there has been an increasing commitment to facilitate communication between deaf or hard of hearing people with people who do not know sign language, but there are still few accessible environments and systems for that. The lack of sign language knowledge in the general population makes communication with deaf people difficult. Thus, the development of automatic translation mechanisms is very important. This work presents new strategies that can help in this process, focused on the identification of individual signs.
Perspectives
In recent decades, speech recognition and text translation between languages have made great strides. In the coming years, this advance should reach sign language recognition.
Dr. Luciano Digiampietri
Universidade de Sao Paulo Campus da Capital
Read the Original
This page is a summary of: Skin Color Segmentation and Levenshtein Distance Recognition of BSL Signs in Video, October 2017, Institute of Electrical & Electronics Engineers (IEEE),
DOI: 10.1109/sibgrapi.2017.19.
You can read the full text:
Contributors
The following have contributed to this page
