Pepster writes: > "Our method is based on a simple premise: neighboring pixels in > space-time that have similar intensities should have similar > colors" > I assume this to mean you need the "timing information of the > stroke", similar to data needed for recognizing handwriting. Nope, it refers to video. I.e. pixel neighbourhood not only in the (x,y) plane, but also in preceeding and succeeding video frames. --tml
