Geometric Computing Lab and Artificial Intelligence Lab, Stanford University
Joint Understanding of 2D Images and 3D Shapes
Monday 11th of April 2016 at 02:00pm
Recent technological developments have led to an explosion in the amount of 3D data. With this background, we build a knowledge-base that stores rich information on top of networks of 3D models. Knowledge stored on ShapeNet can be used as priors of geometry and appearance for understanding 2D images.
We believe that such a 3D-centric knowledge-base, named ShapeNet, is beneficial to many aspects of our life, since we live in a world where 3D is the natural space. ShapeNet is still under construction, targeting at including millions of 3D models from thousands of object classes with semantical, geometrical and physical relationships and attributes. ShapeNet is especially important for solving computer vision problems. We jointly analyze 3D models and 2D images for various applications, including single-image depth estimation, novel view feature synthesis, rendering data for CNN training, and shape-centric image representations by joint embedding.
Hao Su is currently a Ph.D student in the Computer Science Department of Stanford University, advised by Prof. Leonidas Guibas, co-advised by Prof. Silvio Savarese. He is a member of Geometric Computing Lab and Artificial Intelligence Lab. He has also received a Ph.D in Applied Mathematics from Beihang University, specialized in statistics.
Hao's research interest includes computer vision, machine learning and geometry processing. In particular, he is interested joint analysis of 2D images and 3D shapes. His recent work focuses on data-driven shape analysis and information transportation between image and shape world. He has publications in CVPR, ICCV, NIPS, ICML, SIGGRAPH, SIGGRAPH Asia, VLDB, SIGSPATIAL and etc.
Prior to joining Prof. Guibas' group, he was advised by Prof. Fei-Fei Li from 2008 to 2012, advised by Dr. Harry Shum and Dr. Jian Sun at MSRA from 2005 to 2008.
Join Email List
You can subscribe to our weekly seminar email list by sending an email to
firstname.lastname@example.org that contains the words
subscribe redwood in the body of the message.
(Note: The subject line can be arbitrary and will be ignored)