Title: Finding the Right Representations for Generative Modeling of 3D Shapes
Time：10:00-12:30 AM, 22 April ,2019
Unlike images and video, 3D shapes are not confined to one standard representation.This is one of the challenges we face when developing deep neural networks (DNNs) to learn generative models of 3D shapes or virtual scenes. Up to now, voxel grids, multi-view images, point clouds, and integrated surface patches have all been considered. In this talk, I show that traditional convolutional neural networks (CNNs) operating on pixels/voxels may not be best suited for the task. I present our recent works on using implicit shape representations and a generative autoencoder for shape structures to improve the quality of shape generation using DNNs. The latter approach decouples coarse and fine-grained learning of structured data, which is applicable to both indoor scenes and digital documents.
Hao (Richard) Zhang is a full professor in the School of Computing Science at Simon Fraser University (SFU), Canada, where he directs the graphics (GrUVi) lab and is currently the Associate Director of Research and Industrial Relations, as well as the inaugural director of the Professional Masters Program in Visual Computing. He also holds or has held visiting professor positions at Stanford University, Shenzhen University, and Beijing Film Academy. Richard obtained his Ph.D. from the Dynamic Graphics Project (DGP), University of Toronto, and M.Math. and B.Math degrees from the University of Waterloo, all in computer science. His research is in computer graphics with special interests in geometric modeling, analysis and synthesis of 3D contents (e.g., shapes and indoor scenes), machine learning (e.g., generative models for 3D shapes), as well as computational design, fabrication, and creativity. He has published more than 120 articles on these topics, including 50 papers from SIGGRAPH (+Asia) and ACM Trans. on Graphics, the top venue in the field. Richard served as editor-in-chief for Computer Graphics Forum (2014-2018) and an associate editor of several journals includings IEEE Trans. on Visualization and Computer Graphics, IEEE Computer Graphics & Applications, among others. He has served on the program committees of all major computer graphics conferences including SIGGRAPH (+Asia), Eurographics, Symposium on Geometry Processing (SGP), and is SIGGRAPH Asia 2014 course chair, a paper co-chair for SGP 2013, Graphics Interface 2015, and CGI 2018, and a program co-chair for the International Geometry Summit 2019 and SIAM GD 2019. Richard is an IEEE Senior Member and his awards an NSERC DAS (Discovery Accelerator Supplement) Award in 2014, Best Paper Awards from SGP 2008 and CAD/Graphics 2017, a Faculty of Applied Sciences Research Excellence Award at SFU in 2014, and a National Science Foundation of China (NSFC) Overseas Outstanding Young Researcher Award in 2015. For his university service, he received the SFU Dean of Graduate Studies Awards for Excellence in Leadership in 2016.
Title: Collaborative augmented reality: technical, collaboration and social issues
Collaborative Augmented Reality (AR) systems enable multiple people to share the same augmented view, with collaborators being either collocated or remote. These systems are currently at a critical point in time as they are soon to become more commonplace. However, AR technology has only recently matured to the point where researchers can focus on the nuances of supporting collaboration, rather than needing to focus on creating the enabling technology. Leading companies such as Apple and Microsoft are racing to bring new and better AR hardware to the market. Among the possible applications, it is widely viewed that collaborative AR is to be one of the killer applications. The possibilities of collaborative AR are tremendous. From the ability to show a proxy of oneself in order to provide the sense of being there, to using AR in remote collaboration around a physical object (e.g., to help fix a technical problem), or simply using AR in a shared game (such as Pokemon Go). In this talk, I will survey existing collaborative AR topics and describe some of the works that I was involved with that study how to better support collaboration in AR as well as look at various social issues surrounding collaborative AR.
Joel Lanir is an associate professor in the Information Systems department at the University of Haifa, Israel, where he leads the human-computer interaction lab. His research interests lie in the general area of human-computer interaction and information visualization, and more specifically mobile and context-aware computing and the design and evaluation of novel technologies such as augmented reality, wearable computing, and more. Dr. Lanir received his PhD in computer science from the University of British Columbia in 2009. His research was supported by the European Union (FP7), Israel Science Foundation, Israel’s chief scientist and more. Joel regularly publishes in top-tier HCI conferences and journals (CHI, CSCW, HCI journal, IJHCS) and has received the CHI best paper among other awards. Joel is currently a visiting professor at the UBC computer science department.