Embodied Visual Perception and Learning: Developments and Prospects
Embodied intelligence highlights how intelligent agents perceive, learn, and make decisions through physical interaction with their environment. As a core capability of embodied intelligence, visual perception and learning play a crucial role in how agents understand their surroundings and exhibit adaptive behavior. This talk will provide an overview of the key developments in embodied visual perception and learning, tracing the field's evolution from early rule-based approaches to modern end-to-end frameworks that integrate deep learning and reinforcement learning. We will explore advances in multimodal perception, active vision, and sim-to-real transfer, and discuss major challenges such as data efficiency, generalization, computational constraints, and semantic understanding. Finally, the lecture will look ahead to future directions, including brain-inspired visual models, embodiment-language co-evolution, and adaptive learning in open environments, and will outline potential applications in robotics, virtual agents, and human-computer interaction.
Date and Time
Location
Hosts
Registration
-
Add Event to Calendar
- Shenzhen University
- Shenzhen, Guangdong
- China
- Building: Auditorium No. 2, Huixing Building
Speakers
Weishi Zheng
Biography:
Weishi Zheng is a Chair Professor under the Changjiang Scholars Program of the Ministry of Education of China, a Royal Society Newton Advanced Fellow (UK), and a Fellow of the International Association for Pattern Recognition (IAPR). He currently serves as the Director of the Key Laboratory of Machine Intelligence and Advanced Computing, Ministry of Education. Prof. Zheng’s research focuses on theory and methods for collaborative and interactive visual analysis, addressing key challenges in human modeling and robotic behavior understanding through visual computing. He has published over 150 papers in top-tier journals and conferences, including CCF Class A venues, Nature sub-journals, and top journals in the Chinese Academy of Sciences Zone 1. He serves on the editorial boards of leading international journals, including IEEE Transactions on Pattern Analysis and Machine Intelligence (T-PAMI) and the Artificial Intelligence Journal (AIJ). Prof. Zheng has led five major national-level research and talent projects, as well as the Guangdong Provincial Distinguished Youth Team Project. His honors include the First Prize of the Natural Science Award from the Chinese Society of Image and Graphics, the First Prize of Guangdong Natural Science Award, and the Second Prize of the National Teaching Achievement Award, among others.