Short Biography
My name is Haomian (Eric) Wang. I currently study at Computer science and engineering Dept., Hong Kong University of Science and technology (HKUST) as PG student. My supervisor is Dr. Huamin Qu, director of Visualization group. I received my Bachelor degree of computer science and technology at Special Class for Gifted Young Dept., University of science and technology of China (USTC).
Research Interests
My research interests include computer graphics, computer vision, computer visualization, and image processing.
Personal CV
Chinese Version English Version
E-mail: whaomian@ust.hk or whaomian@gmail.com
Office: Rm4204, HKUST, Clear Water Bay, Hong Kong
| Year 2007: Excellent BA Thesis of USTC |
| Year 2006: Microsoft Research Asia Young Scholar |
| Year 2005: "Huawei Cup" Software Design Match of USTC Third Class |
| Year 2003: USTC Junior Football match Champion |
-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Haomian Wang, Houqiang Li, Baoxin Li, “Video Inpainting for Largely Occluded Moving Human”, International Conference on Multimedia & Expo (ICME), Beijing, China, July. 25, 2007.
Huamin Qu, Haomian Wang, Weiwei Cui, Yingcai Wu, Ming-Yuen Chan, “Focus+Context Route Zooming and Information Overlay in 3D Urban Environments”, Conditional Accepted by IEEE, Transaction of Visualization and Computer Graphics.
In this project, I set up a 3D digital city system, call CityVis, which can renders building models extracted from KML format file (KML format is used in Google Earth to overlay information). In CityVis, user can modify the 3D buildings, including the sizes, positions, textures and so on. The system is implemented using VTK (Visualization Toolkit). Further research about 3D digital city is carried out in CityVis.
In this project, we propose a novel approach to low bit rate image coding. Our work is inspired by recent progress on image hallucination, which reconstructs a high resolution image from a low resolution image. In our approach, we combine the technique of image super-resolution with JEPG2000 standard for image compression. Experiments demonstrate that our approach can obtain higher visual quality compared with JEPG2000, when compressing images at the same low bit rate.