报告时间：2019年8月19日（星期一）上午10:30
报告地点：浙江大学紫金港校区蒙民伟楼402室
报告题目：Geometry Learning under Hybrid 3D Representations
主讲人：Qixing Huang（University of Texas at Austin）
主持人：周晓巍 研究员

Abstract: Data representations play a central role in machine learning algorithms. While there are canonical data representations for visual data in the form of images and videos, it becomes less clear what are suitable data representations for 3D geometry. In the literature, people have studied representations such as point clouds, triangular meshes, parametric surfaces,  implicit surfaces, and multi-views. However, none of these are ideal for all geometry processing applications. In this talk, I will argue that instead of focusing on a single representation, it is critical to combine multiple representations together as hybrid 3D representations. Specifically, I will show the advantage of hybrid 3D representations in several applications such as relative pose estimation, scene understanding, scene synthesis.

Bio:Qixing Huang is an assistant professor of Computer Science at the University of Texas at Austin. He obtained his PhD in Computer Science from Stanford University. He was a research assistant professor at Toyota Technological Institute at Chicago before joining UT Austin. Dr. Huang's research spans the fields of computer vision, computer graphics, and machine learning, and publishes extensively in venues such as NeuriPS, ICML, CVPR, ICCV, and etc. He is also interested in statistical data analysis, compressive sensing, low-rank matrix recovery, and large-scale optimization, which provides theoretical foundation for his research. He also received the best paper award at the Symposium on Geometry Processing 2013, the best dataset award at the Symposium on Geometry Processing 2018, and the most cited paper award of Computer-Aided Geometric Design in 2010 and 2011. Dr. Huang was an area chair for CVPR 2019 and ICCV 2019. He will serve as a paper co-chair of Symposium on Geometry Processing 2020.