beeline video The video shown resulted from a process that transformed still images shot in Toronto, mainly skylines and close-ups of small objects found while walking. Two video systems followed an activity sequence as follows: The projected image fades to white, and four images and math operations for the red, green, and blue channels are selected at random. The images are crossfaded in pairs at different rates, and the resulting streams fed into the inputs of an object applying the selected math operations. The projected image fades back from white to show the created sequence, which runs for about four minutes until fading to white again. In order for the process to run in real time, the images were 320 x 240 and scaled to full size. As a result certain combinations of operators yielded very active internal pixellation that varied from smooth motion to chaotic transitions. The random combination of math operators determined which parts of the images were combined or became keying masks for other parts.