In a smart cyberspace, visual contents such as images, videos and 3D models have been demonstrating impressive impact on our daily life. However, such visual contents, especially the interactable 3D models, are still far from sufficient to meet the individual requirements of billions of cyberspace users. In this talk, I will present our recent research efforts on data-driven visual content understanding and production, i.e., the large-scale generation of visual contents by using the knowledge mined from data. In particular, I will focus on two main issues: the role of data in computer vision and computer graphics, and the difference between laboratory generation and large-scale production.
Biography: Xiaowu Chen is currently a full professor at State Key Laboratory of Virtual Reality Technology & Systems, School of Computer Science & Engineering, Beihang University. His research interests include virtual reality, augmented reality, visual computing, computer graphics, computer vision, computational photography, human-computer interaction, digital media content, and big data. He is particularly interested in developing data-driven theories, algorithms and systems for intelligent processing and understanding of visual data. His recent works were published in several journals and conferences such as TOG, TPAMI, TIP, CVPR, CGF, CVIU, PR and ECCV.