On-site AI Acceleration Challenge at ICCV 2019



We would like to bring awareness to the energy efficiency of AI accelerators and encourage researchers to innovate a new neural network architecture optimized for AI accelerators.

The challenge is to develop a realtime image classification model that runs on target HW platforms.

We have two tracks: DSP and FPGA. Cash awards will be given to the top three teams in each track. A participant (or a team) can submit to and win prizes in either or both categories.

Please visit the following websites to find more details.

Participants need to send their final submission by 9 AM on the 28th of October. We will announce the top three teams at the end of the Low-Power Computer Vision workshop at 5 PM.



  • DSP track
    • Jaeyoun Kim <jaeyounkim at google.com>
    • Bo Chen <bochen at google.com>
    • Achille Brighton <aib at google.com>
    • Paul Roberts <pwroberts at google.com>
  • FPGA track
    • Naveen Purushotham <npurusho at xilinx.com>
    • Hong Luo <hongluo at xilinx.com>
    • Ashish Sirasao <asirasa at xilinx.com>