Computer Vision

• Core Domains:

• Image Enhancement and Restoration: Designing lightweight deep learning models (e.g., Variant-Depth Networks) for real-time deblurring and super-resolution reconstruction in traffic surveillance and medical imaging.

• Object Segmentation and Understanding: Developing precise algorithms for vascular and organ segmentation to enhance surgical navigation systems in robot-assisted interventions.

• Multimodal Perception: Exploring cross-modal content generation and understanding by integrating visual and textual data.

• Highlights: Research published in IEEE Transactions on Intelligent Transportation Systems (IF=9.5), Journal of Systems Architecture, and 2 authorized national invention patents.

Research Features

• Interdisciplinarity: Combines computer science, art, and pedagogy, emphasizing both technical rigor and humanistic values.

• Cutting-Edge Technologies: Supported by National Natural Science Foundation projects, focusing on diffusion models, lightweight networks, and real-world problem-solving.

• Industry-Academia Collaboration: Partners with companies like Zhuhai Vision Technology to industrialize AI algorithms in smart healthcare and industrial design.

Join the Research Group to Engage in:

• National-level projects exploring frontier AI models and vision algorithms;

• Cross-border collaborations with institutions like Hong Kong Polytechnic University;

• End-to-end research from theory to application, with opportunities for high-impact publications or patents.

Together, Let’s Shape the Future of AI-Powered Art and Vision Science!