Deep generative models for images and 3D shapes

Traditionally, computer vision focuses on image analysis and recognition, and the major interest in the computer graphics community is to produce improved or interesting imagery. In recent years however we have seen a significant trend that computer vision and computer graphics are merging.  The boundary between them becomes increasingly vague and both benefit significantly from recent development in machine learning. In this talk, I will give an overview of our recent research in this joint area which aims to develop automatic methods to generate novel visual content, including images and 3D shapes, using deep generative models. In particular, we focus on tasks which are traditionally challenging, including artistic image stylisation and 3D shape generation.  

Short biography

Dr Yukun Lai is a Reader in the Visual Computing group, School of Computer Science and Informatics, Cardiff University, UK. He obtained his PhD degree from Tsinghua University, China in 2008 and joined Cardiff University shortly after that. He has been working in broad areas of Visual Computing, including Computer Graphics, Geometric Processing, Image Processing and Computer Vision. He has published over 150 papers in journals and conferences, including ACM TOG, IEEE TPAMI, TVCG and TIP, as well as leading conferences such as CVPR/ICCV. He is on the editorial boards of Computer Graphics Forum and The Visual Computer, conference co-chair of SGP 2014 and CVM 2016, BMVC 2018-2020 Area Chair, and on the programme committee of many major international conferences.