Research
- Deep learning for human motion enhancement
- Deep learning for scene labeling
- Manifold learning for human motion modeling
- First-person and egocentric vision
- 3D indoor scene modeling
- Human pose estimation
- Target tracking and recognition
- Sports video mining
- Multi-view face recognition
Signal Image and Video processing
- Modeling and segmentation
- Denoising and enhancement
- Image super-resolution
- Image compression
Biomedical Imaging and Remote Sensing
- 3D retinal curvature estimation
- Feature extraction for retinal images
- Retinal image registration
- Multi-source data analysis and classification
- Web-GIS decision support system (DSS)