RESEARCH INTERESTS: 

My research interests mainly include Machine Learning, Multimedia Computing, Computer Vision and Pattern Recognition. I am particularly interested in the following topics: 

Current topics:  

(1) Deep Learning for Low-Level Vision: New Deep Architectures and Models for Degraded Image/Video Restoration (e.g., denoising, deraining, deblurring and dehazing), Low-Light Image Enhancement, and High-Level Task Driven Low-Level Vision, with emerging applications (e.g., autonomous driving, visual imaging systems, video surveillance and robot vision) in complex environments.  

(2) Image/Video Generation and Deepfake Detection: New Deep Architectures and Models for Text-to-Image Generation, Text-to-Video Generation and Image-to-Video Generation, with emerging real-world applications (e.g., digital human and secure AI).  

(3) Model Compression and Deployment of DNNs: New Network Quantization Methods or Lightweight Architectures for High-level and Low-level Vision Tasks, with deployment on resource limited edge devices (e.g., mobile devices, robot vision and visual imaging system).  

Previous topics (still ongoing): 

(4) Low-Dimensional Modeling of High-Dimensional Data: Sparse Dictionary Learning, Low-Rank Coding, Concept Factorization, Manifold Learning, and Their Deep/Multilayer Extensions, for feature learning and extraction.   

(5) Semi-Supervised Classifier Modeling: Graph based Novel Label Propagation Algorithms, and Deep Semi-Supervised Learning, with application to image annotation and classification. 


PROJECT & FUNDING: 


AWARD & HONOR: 


COLLABORATORS: 

 

RESOURCE LINKS: