The 2020 Embedded Deep Learning Object Detection Model Compression Competition for traffic in Asian countries held in IEEE ICME2020 focuses on the object detection technologies in autonomous driving scenarios. This competition targets on realizing the deep learning object detection through embedded systems with low complexity and model size. The dataset, recorded in Asian countries (e.g., Taiwan), contains several harsh driving environments, including crowded scooters, bicycles, pedestrians and vehicles. There are 89, 002 annotated images provided for training and 1, 000 images for validation. Additional 5, 400 testing images are used in the contest evaluation process, in which 2, 700 of them is used in the qualification stage, and the rest are used in the final stage. There are in total 133 of registered participating teams joining this competition. We have selected the top 10 teams with the highest detection accuracy entering the final competition, from which 8 teams submitted the final results. The overall best model belongs to team USTC-NELSLIP, followed by team "BUPT_MCPRL"and team "DD_VISION". Two special awards of best bicycle and scooter detections go to teams "IBDO-AIOT"and "Deep Learner", respectively.