Mode: Mixed
Embodied photography and cinematography require a robot to integrate aesthetic judgment, camera control, and scene understanding in real time. This talk presents a unified view of the problem, spanning how robots can learn photographic style from human demonstrations, how they can robustly control focus and exposure in challenging lighting conditions, and how these capabilities can enable new applications such as intelligent birdwatching. We first discuss imitation learning methods that capture photographer intent and composition. We then cover event-based autofocus and auto-exposure systems that maintain image quality under low light and extreme illumination. Finally, we explore open-world tracking and language-guided observation for targeted wildlife capture. Together, these directions suggest a future in which robots act as active visual partners, not just passive imaging devices.
Professor Jia Pan is an Associate Professor in the Department of AI and Data Science at the University of Hong Kong. His research spans robotics, embodied intelligence, motion planning, and learning-based control, with applications in navigation, manipulation, and human-centered robot systems. He earned his Ph.D. in Computer Science from UNC Chapel Hill and has published extensively in top robotics venues.
Clean water and clean air are vital for public health. This project focuses on developing high-efficiency and environmentally sustainable filters for removing harmful air/water pollutants. The team has developed novel architectures and functionalities for the filters to achieve high permeance, high removal efficiency, and excellent reusability.
