Image Deblurring


Since my Ph.D. study, I've been trying to make image deblurring more practical solution for consumer photography. Specifically, I've worked on blind deconvolution, non-blind deconvolution, noise & outlier handling, and non-uniform deblurring. Especially, my efficient blind deconvolution method, which was presented at SIGGRAPH Asia 2009, has been proven to be one of the fastest and most reliable methods by several following works by other researchers, and it is still one of the state-of-the-arts.

For the complete list of my published papers, please refer to Publications.

When I was a postdoc at Adobe Research, I had worked on a tech transfer project, which aimed at shipping deblurring technology with Adobe Photoshop. The first version of "Shake Reduction", which was based on my research, was first released in the summer of 2013 as a new feature of Photoshop CC.

Photoshop CC: Camera Shake Reduction


Below is a demo presented at Adobe MAX 2011 Sneak Peek, which sparked huge buzz on the internet.

Adobe MAX 2011 Sneak Peek



Video Restoration

I am also interested in video restoration problems. I developed a highly efficient and robust video deblurring algorithm, which doesn't rely neither on accurate blur estimation nor deconvolution, and presented it at SIGGRAPH 2012.

Below is a demo video of my video deblurring. Because of video compression of Youtube, it's not easy to clearly see the effects of video deblurring. For better video, please visit its project webpage. You can find other details there.

Video Deblurring



Past Projects

Development of a 3D game engine for mobile environments
This project was conducted by POSTECH Computer Graphics Lab. and Sinjisoft to make a C++ based 3D game engine for mobile environments such as cell phones. The developed game engine fully supports JSR-184 spec. I participated in this project from Sept. 2005 to Feb. 2006. I mainly implemented the animation related parts of the engine and an exporter plugin for 3DS Max.
Oriental Painterly Rendering
I studied this subject in the CS499 undergraduate research course in 2004. In this project, I made an algorithm that produces an oriental black ink drawing from a picture. [PDF (in Korean)]