Issue |
ITM Web Conf.
Volume 70, 2025
2024 2nd International Conference on Data Science, Advanced Algorithm and Intelligent Computing (DAI 2024)
|
|
---|---|---|
Article Number | 02010 | |
Number of page(s) | 8 | |
Section | Machine Learning in Healthcare and Finance | |
DOI | https://doi.org/10.1051/itmconf/20257002010 | |
Published online | 23 January 2025 |
Overview of Sign Language Translation Based on Natural Language Processing
Software Department, Henan University, 475004, Kaifeng, China
Corresponding author: 2225050185@henu.edu.cn
This paper explores the progress, challenges, and future directions in Sign Language Translation (SLT) within the broader field of Sign Language Processing (SLP), which combines Computer Vision (CV) and Natural Language Processing (NLP) to translate sign language videos into spoken language texts. The study begins by examining various sign language representation methods, such as video, gesture, symbol systems, and annotation, analyzing their strengths and weaknesses. It highlights the critical need for high-quality, large-scale datasets to advance SLT research, while acknowledging challenges like data scarcity, annotation inconsistencies, and ethical concerns. The paper then reviews recent SLT research, identifying key challenges and proposing solutions, such as expanding datasets through collaboration with the deaf and hard-of-hearing community, and employing advanced data collection techniques. Additionally, it suggests applying NLP methods like transfer learning and large language models to address specific challenges. Finally, the paper advocates for stronger interdisciplinary collaboration between CV and NLP to develop models and algorithms that are better suited to the unique aspects of sign languages.
© The Authors, published by EDP Sciences, 2025
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.