The Vitis Vision library is a set of select OpenCV functions optimized for Zynq-7000, Zynq UltraScale+ MPSoC, Versal VCK190, Alveo U200, and U50 devices. The maximum resolution supported for all the functions is 4K, except Houghlines and HOG (RB mode).
Important
All functions in the library are implemented in the streaming model except 7. Bounding box, Canny, Cornertracker, Crop, EdgeTracing, MeanShiftTracking, Rotate are memory mapped implemenations. These functions need to have the flag __SDA_MEM_MAP__ set for compiling correctly
Important
The default depth value for all the memory mapped implementations (Bounding box, Canny, Cornertracker, Crop, EdgeTracing, MeanShiftTracking, Rotate) is “_XFCVDEPTH_DEFAULT = -1”. The default depth value for all the streaming model implementations is “_XFCVDEPTH_DEFAULT = 2”.
Note
The Number of pixel per clock depends on the maximum bus width a device can support. For example: Zynq-7000 SoC has 64-bit interface and so for a pixel type 16UC1, maximum of four pixel per clock (XF_NPPC4) is possible.