Close this search box.

Taking Acceleration to the Edge

Optical Flow is a technique to estimate the apparent motion of image objects caused by their movements in a video sequence.

Field Programmable Gate Arrays (FPGAs) are ideally suited for optical flow. Data collected can be used for pattern recognition, object sorting, robotic arm control, and more.

FPGAs can also be used as a vision processing accelerator inside the Edge Computing platform, to harness the power of artificial intelligence deep learning for analysis of the Optical Flow data.

"Acceleration is the path forward for computing. The tremendous result shown by using an FPGA opens up a whole new world of computing power!"

AimValley Optical Flow Demonstrator

AimValley’s Optical Flow Demonstrator enables comparison of implementations running in software on a CPU and offloading with a graphics card and FPGA accelerator card. The design reads from a video file or live images from a webcam and writes outputs directly to a display or to a file.

The AimValley Optical Flow Demonstrator combines the Lucas Kanade Algorithm with Harris Corner Detection and color wheel encoding

System Configuration

  • Standard x86 server PC
  • FPGA card
  • GPU card
  • Open CV, OpenCL, C/C++ and HLS library components

Measurements Results

Ask for a demo



AimValley - Optical Flow Demonstrator - How can we help you?

Need Help?

WE'RE Here To Assist You

Feel free to contact us, and we will be more than happy to answer all of your questions and provide a demo.