Socially-aware Artificial Intelligence for Last-mile Mobility

This lecture covers the research on socially-aware artificial intelligence for last-mile mobility, focusing on understanding social etiquettes, anticipating behaviors, and forecasting crowd movements using deep learning models. The instructor presents the challenges in computer vision for transportation, such as limited resolution, safety-critical accuracy, and real-time efficiency. The lecture also delves into the development of a privacy-safe social distancing tool based on visual intelligence, collaborative sampling in generative adversarial networks, and the universal representation learning framework for joint perception, social forecasting, and acting.