Self-Annotating Attention
George discusses the challenges of labeling human attention in driver monitoring systems, emphasizing the need for human supervision in annotation. He suggests that self-annotating techniques, such as analyzing steering wheel movements, could provide insights into attention lapses. As the team grows, exploring these self-supervised methods will become a priority.In this clip
From this podcast

Lex Fridman Podcast
George Hotz: Hacking the Simulation & Learning to Drive with Neural Nets | Lex Fridman Podcast #132
Related Questions