VoC: a reconfigurable matrix for stereo vision processing

This paper presents a reconfigurable matrix VoC that can be applied to stereo vision computation. VoC accelerates block pixel matching by providing a highly parallel implementation of the sum of absolute differences metric. Reconfigurability allows VoC to deal with different block sizes, ranging from a single 7 /spl times/ 7 SAD computation to 9 simultaneous 5 /spl times/ 5 block computations. The pipelined version mapped to Xilinx FPGA could be simulated at 158 MHz, producing 1.42 billion matching per second.