TeeVR: Revolutionary VR Content Generation from the Real World


November 21, 2017

Monday, November 20, 2017
Dr. Nakju Doh
Pohang University of Science and Technology
11:00am - 12:00pm
SEH, 2990

 

Abstract

In this presentation, a revolutionary framework for VR content generation from the real world, called 'TeeVR', is introduced which has four special features. First, TeeVR can scan 3D real world by just walking around the target space with a special back-pack type sensor system. Second, most of the map building procedures are automatic. Third, the coverage of spaces span from small houses to large-complex mall as well as indoor and outdoor seamlessly. Lastly, not the least, TeeVR is the unique VR content which provides high resolution image (more than Full-HD) from the real-world in any user's view-point. Its technical pipelines consist of 3D SLAM from RGB-D point clouds, automatic segmentation of PCD in object-wise manner, automatic mesh generation, and the realistic image rendering.

 

Biography

Nakju Doh received the BS, MS, and PhD degrees from Pohang University of Science and Technology (POSTECH), Korea, in 1998, 2000, and 2005, respectively. From 2005 to 2006, he worked as a Senior Research Scientist with the Robotics Division, Electronics and Telecommunications Research Institute. Since 2006, he has been a Professor with the School of Electrical Engineering, Korea University, Seoul, Korea. His research interests include realistic modeling of indoor-spaces, localization problem of mobile robots, 3-D SLAM, semantic mapping, sensor fusion of range and image data, robot motion planning, etc. Dr. Doh was the recipient of the best Student Paper Awards at the IEEE International Conference on Robotics and Automation in 2003.