エンドツーエンド自動運転データセットに潜むバイアス
Hidden Biases of End-to-End Driving Datasets
CARLA Leaderboard 2.0向けのエンドツーエンド自動運転において、アーキテクチャではなく訓練データセットを体系的に分析し、専門家の運転スタイルやフレーム選択基準の重要性を明らかにした。
著者: Julian Zimmerlin, Jens Beißwenger, Bernhard Jaeger, Andreas Geiger, Kashyap Chitta
分類: cs.CV, cs.AI, cs.LG, cs.RO
原文アブストラクト
End-to-end driving systems have made rapid progress, but have so far not been applied to the challenging new CARLA Leaderboard 2.0. Further, while there is a large body of literature on end-to-end architectures and training strategies, the impact of the training dataset is often overlooked. In this work, we make a first attempt at end-to-end driving for Leaderboard 2.0. Instead of investigating architectures, we systematically analyze the training dataset, leading to new insights: (1) Expert style significantly affects downstream policy performance. (2) In complex data sets, the frames should not be weighted on the basis of simplistic criteria such as class frequencies. (3) Instead, estimating whether a frame changes the target labels compared to previous frames can reduce the size of the dataset without removing important information. By incorporating these findings, our model ranks first and second respectively on the map and sensors tracks of the 2024 CARLA Challenge, and sets a new state-of-the-art on the Bench2Drive test routes. Finally, we uncover a design flaw in the current evaluation metrics and propose a modification for future challenges. Our dataset, code, and pre-trained models are publicly available at https://github.com/autonomousvision/carla_garage.
