My research into computer vision techniques began as a final year undergraduate project with Bob Fisher at the University of Edinburgh. I went on to do a PhD at the University of Surrey, where I worked on developing free-viewpoint video systems using various technologies. I am currently working on an eclectic mix of topics related to computer vision at the University of British Columbia.
Current research areas:
- Architecture for distributed vision processing
- Creating a viable vision system capable of being accelerated in hardware, and being easy to use by vision researchers and hobbyists.
- Multiple view vision algorithms (synthesis, tracking)
- Mac OS X (Xcode), Windows (Visual Studio), Symbian, linux (when I have to)
- C++, Python, C#
- OpenGL, GLSL, CUDA
- 3D reconstruction of dynamic scenes using video from multiple views
- Free viewpoint systems to view recorded scenes as 3D video
- Efficient and exact visual hull construction
- Surface refinement via photo consistency and stereo
- Architectural object reconstruction and completion from range data