The scientific topics of VAS are centered around the theme of perception at scale. We focus on deep learning-based perception for autonomous driving, especially on scaling existing visual perception models to novel domains (e.g. adverse weather/lighting conditions, low-quality data),  to more data modality (e.g. Radar, HD Maps), to unseen classes (e.g. rare classes), and to more tasks. Our research topics are summarized in the diagram below.