Demonstration: Implementation of three mainstream machine learning algorithms (CNN, dense optical flow, stereo vision)

Computer vision, machine learning, and sensor fusion in a single chip using the reVISION stack

This video demonstration combines 3 of the most common complex algorithms used in today's "vision-guided systems", including Convolutional Neural Networks (CNN) for object detection or scene segmentation, Dense Optical Flow for motion tracking , and Stereo Vision for depth perception , all running on a single Zynq Ultrascale+ MPSoC device.