Adaptive Census Transform: A novel hardware-oriented stereovision algorithm

This paper presents a new hardware-oriented approach for the extraction of disparity maps from stereo images. The proposed method is based on the herein named Adaptive Census Transform that exploits adaptive support weights during the image transformation; the adaptively weighted sum of SADs is then used as the dissimilarity metric. Quality tests show that the proposed method reaches significantly better accuracy than alternative hardware-oriented approaches. To demonstrate the practical hardware feasibility, a specific architecture has been designed and its implementation has been carried out using a single FPGA chip. Such a VLSI implementation allows a frame rate up to 68 fps to be reached for 640 × 480 stereo images, using just 80,000 slices and 32 RAM blocks of a Virtex6 chip.

► We propose a new hardware-oriented approach for the extraction of disparity maps.
► Adaptive support weights are unconventionally exploited during the image transformation.
► Accuracy tests have been performed and demonstrate the effectiveness of the design choices.
► An hardware architecture has been designed and implemented using a single FPGA chip.
► The proposed architecture reaches a good trade-off between speed-performance and accuracy.