Presentation for explaining the paper "Is Ego Status All You Need for Open-Loop End-to-End Autonomous Driving?" presented from Nanjing University and NVIDIA.
In recent years, end-to-end methods using DNNs have been studied for autonomous driving systems, and have demonstrated high performance on datasets such as the nuScenes. However, this paper raises the question of whether this high performance is truly due to high perception, prediction, and planning.
"Is ego status all you need for achieving high scores?"