|
|
Byung Tae
Oh |
Byung Tae Oh received his B.S. degree in Electrical Engineering at Yonsei University, Seoul, Korea, in 2003, and his M.S. and Ph.D. degree in Electrical Engineering at University of Southern California, LA, CA in 2007 and 2009, respectively. He was a member of Media Communication Group advided by C.-C. Jay Kuo with working image and video restoration, super-resolution, analysis/synthesis and pre/post-processing for coding.
Objective
- Full-time position for R&D in the area of image/video processing
Research Interests
- Image and video coding / compression
- Image super-resolution
- Texture and noise analysis and synthesis
Education
- Ph.D. Aug. 2009
Department of Eletrical Engineering, University of Southern California, USA - Dissertation : Texture processing for image/video coding and super-resolution applications
- Advisor: C.-C. Jay Kuo
- Master of Science, May 2007
Department of Eletrical Engineering, University of Southern California, USA - Bachelor of Engineering, Aug. 2003
Department of Eletrical Engineering, Yonsei University, Seoul Korea
Experience
- Research Assistant, Prof. C. -C. Kuo, University of Southern California, Aug. 2005 ~ Aug. 2009
- Image retrieval for bill counterfeit protection algorithm
- Film grain noise modeling algorithm for advanced video coding
- Synthesis-based texture video and image coding algorithm
- Stochastic texture image interpolation / super-resolution
- Internship in Sharp Lab. of America, May~Aug. 2006, May~Aug. 2007
- Develop the film grain noise modeling and rendering algorithm
- Develop the synthesis-based texture video and image coding algorithm
- Undergraduate Research, Prof. M. G. Kang, Yonsei University, 2003
- Finding fundamental matrix for self-calibration in stereo image processing
Research
- Stochastic texture image super-resolution, May 08 ~ Aug. 09
- Develop the stochastic-texture oriented super-resolution algorithm
- Find more accurate auto-regressive model parameters by self-learning scheme
- Apply non-local scheme to refine and update model parameters
- Derive the objective function based on the probability model with EM algorithm
- Synthesis-based texture video and image coding algorithm, May 07 ~ Apr. 08
- Improve coding efficiency by replacing conventional coding with texture synthesis technique
- Send user-defined side-information to control the amount of synthesis
- Optimally assign locally adaptive amount of side information
- Analyze and evaluate synthesis-based coding approach with quantitative method
- Film grain noise modeling in advanced video coding, May 06 ~ Apr. 07
- Improve coding efficiency by pre- and post-processing for film grain noise
- Extract and parameterize film grain noise using specific properties of film grain
- Build a model to synthesize film grain noise with objective quality criteria
Publications
- B. T. Oh and C.-C. Kuo, ˇ°New PAR/NL Scheme for Stochastic Texture Super-Resolutionˇ±, IEEE Trans. on Image Processing, in preparation.
- B. T. Oh, Y. Su, A. Segall and C.-C. Kuo, ˇ°Synthesis-based texture video coding with side informationˇ±, IEEE Trans. on Circuit and System for Video Technology, submitted.
- B. T. Oh, S. Lei and C.-C. Kuo, ˇ°Film grain noise analysis and synthesis for high definition video codingˇ±, IEEE Trans. on Circuit and System for Video Technology, accepted.
- B. T. Oh and C.-C. Kuo, ˇ°New PAR/NL scheme for stochastic texture interpolationˇ±, in Proceedings of IEEE International Conference on Multimedia Expo, 2009.
- B. T. Oh, Y. Su, A. Segall and C.-C. Kuo, ˇ°Synthesis-based texture video coding with side informationˇ±, in Proceedings of IEEE International Conference on Image Processing, 2008.
- B. T. Oh, S. Sun, S. Lei and C.-C. Kuo, ˇ°Film grain noise modeling in advanced video codingˇ±, Proceedings of SPIE, Visual Communications and Image Processing, 2007.
Journal Papers
Conference Papers
Last Update : Aug. 1. 2009