One of main recent computing trends is that CPU and GPU improve their performance by increasing the number of cores on a chip instead of increasing clock frequency. As a result, it is critical to utilize such multi-cores to handle large-scale models and get benefits from hardware advance.
Particularly, we have been working on designing various techniques utilizing heterogeneous computing resource including CPUs and GPUs. It is very important to respect different characteristics of CPUs and GPU, and to design algorithms given those characteristics.
High Performance Graphics (HPG), 2017
ICRA2017 Workshop on Robotics and Vehicular Technologies for Self-driving cars, 2017
Source code: ZIP file,
Github page
Optics Express, 2015
Vol. 23, Iss. 3, pp. 2101-2120
High Performance Graphics, 2014
IEEE Transactions on Visualization and Computer Graphics (TVCG), 2013
IEEE Transactions on Visualization and Computer Graphics(TVCG), 2013
(the Spotlight Paper for the September issue)
IEEE Transactions on Very Large Scale Integration Systems (VLSI), 2012
Computer Graphics Forum (Pacific Graphics), 2009
Received a distinguished paper award at the conf.
TopACM SIGGRAPH, 2003