MONDAY July 20, 3:30pm - 5:00pm
TOPIC AREA: DESIGN
KEYWORD: CONTEST
EVENT TYPE: DESIGNER TRACK
When Accuracy Meets Power: 2020 DAC System Design Contest on Low Power Object Detection
Organizers:
Jeff Goeders - Brigham Young Univ., Provo, UT
Jingtong Hu - Univ. of Pittsburgh, Pittsburgh, PA
Philip Brisk - Univ. of California, Riverside, CA
Naveen Purushotham - Xilinx Inc., San Jose, CA
This special session highlights the winning entries of the 2020 DAC System Design Contest on Low Power Object Detection (SDC). The contest started early this year. Contestants were required to implement object detection machine learning algorithms in FPGA to achieve high accuracy and low power. With Xilinx’s sponsorship, contestants competed in the category FPGA using Xilinx Ultra 96. All contestants used a large training dataset provided by DJI, a company renowned for drone technologies. The dataset contains over 14 GB of images with 100 different objects to detect. A hidden dataset was used to evaluate the performance of the designs in terms of accuracy and power. This year’s contest attracted over 80 teams from more than 10 countries/regions. In this session, the SDC organizers will first introduce the background and statistics for this year’s contest and announce the competition results. Then the top three teams in the FPGA category will present their designs.

Thank you to our 2020 System Design Contest Sponsors: