Colour Transfer By Feature Based Histogram Registration
A common problem in computer vision is that different sensors acquire different colour responses to an imaged object. This problem occurs because physical factors during the imaging process introduce a variation that differs for each sensor; in addition, it is practically impossible to image an object under perfectly constant lighting conditions at different spatial positions within an imaging environment. This variation degrades the performance of colour computer vision processes such as object tracking; in addition, the involved nature of calibration routines means that the calibration step is often ignored.
Publication: Proceedings of the British Machine Vision Conference 2007. Malvern, UK: British Machine Vision Association | full text (PDF)
Year: 2007
My background is in Computer Science and Machine Learning, and I plan to apply Machine Learning and probabilistic methods to long-standing vision problems. In particular, I'm looking at the problem of how to reconstruct a 3D model from video footage of the interior of a building.