Research Topics
Get to Know Us
This is a paragraph about your business. Let your visitors know who you are, what you do and what your website is all about. Double click on the text box to edit the text and add all the information you want to share.
Our
Story
Sign Language
Retrieval
We focus on sign language understanding and retrieval, an emerging research area that integrates deep learning with computer vision and natural language processing. This field aims to interpret the visual-gestural communication systems used by deaf communities around the world. Core tasks include sign recognition (detecting individual signs from video), sign language translation (converting signed content into text or speech), and cross-modal retrieval (linking sign language videos with corresponding text queries and vice versa). These tasks involve distinct challenges, such as capturing the spatial-temporal dynamics of signing, accounting for non-manual signals, and aligning multi-modal inputs to bridge visual gestures with linguistic meaning.
This research explores the design and analysis of genetic algorithms for solving complex optimization problems, particularly in multi-modal and multi-objective settings.
By drawing on principles of natural evolution, such as selection, crossover, and mutation, genetic algorithms offer robust solutions in search spaces with multiple local optima and conflicting objectives. Current work emphasizes enhancing algorithm performance through techniques such as clustering, adaptive parameter control, and probabilistic modeling, with a strong foundation in mathematical modeling and theoretical analysis.