3D Vision and Beyond
Vision is essential in the development of intelligence and life. This talk introduces our works in 3D computer vision and their applications. We first introduce a novel ‘global’ structure-from-motion (SFM) algorithm to solve camera poses, which leads to better accuracy and efficiency and scales up to millions of cameras. Next, we will solve dense per-pixel depth by maximizing photo-consistency, and develop learnable scene priors and learn-to-optimize techniques for improved accuracy and efficiency. We then introduce some recent works in non-rigid scene reconstruction including applications in face and human body reconstruction. At last, some applications in image/video editing, robotics, AR/VR will be presented.
Dr. Ping Tan is an associate professor at the Simon Fraser University (SFU) and the director of Alibaba AI Lab (while on leave from SFU). Before that, he was an associate professor at the National University of Singapore (NUS). He obtained his Ph.D. degree from the Hong Kong University of Science and Technology (HKUST) in 2007 and MSc and BSc degrees from the Shanghai Jiao Tong University in 2003 and 2000, respectively. He serves on the editorial board of the major computer vision and graphics journals, including IEEE TPAMI, IJCV, CGF, MVA, and serves as an area chair for major computer vision and graphics conferences, including CVPR, ICCV, SIGGRAPH, SIGGRAPH Asia, IROS, 3DV, AAAI.