The 4th ACM SIGGRAPH Conference and Exhibition on Computer Graphics and Interactive Techniques in Asia
Conference 12-15 December • Exhibition 13-15 December • Hong Kong Convention & Exhibition Centre
 

Technical Papers

Image Processing

Thursday, 15 December 11:00 - 12:45 |  Convention Hall B

Session Chair
CK Tang

Image Smoothing via L0 Gradient Minimization - Picture

Image Smoothing via L0 Gradient Minimization


A new image smoothing method that can globally maintain high-contrast structures and sharpen the most prominent edges while suppressing low-amplitude details. It is achieved in a new global optimization procedure and is particularly useful in edge representation and extraction as well as other enhancement and manipulation tasks.


Li Xu, The Chinese University of Hong Kong
Cewu Lu, The Chinese University of Hong Kong
Yi Xu, The Chinese University of Hong Kong
Jiaya Jia, The Chinese University of Hong Kong


Convolution Pyramids - Picture

Convolution Pyramids


We present a new approach for rapidly approximating convolutions with large kernels. We demonstrate that our method outperforms existing state-of-the-art methods, and is well suited for tasks such as gradient field integration, seamless image cloning, and scattered data interpolation.


Zeev Farbman, The Hebrew University
Raanan Fattal, The Hebrew University
Dani Lischinski, The Hebrew University


GPU-Efficient Recursive Filtering and Summed-Area Tables - Picture

GPU-Efficient Recursive Filtering and Summed-Area Tables


We present innovative state-of-the-art parallel algorithms for the combined evaluation of recursive filters over rows and columns of images. These can be used, for example, to solve the B-spline interpolation problem or to compute summed area tables. We report significant speedups in graphics processing unit (GPU) computation.


Diego Nehab, National Institute for Pure and Applied Mathematics
André Maximo, National Institute for Pure and Applied Mathematics
Rodolfo Lima, National Institute for Pure and Applied Mathematics
Hugues Hoppe, Microsoft Research


Multigrid and Multilevel Preconditioners for Computational Photography - Picture

Multigrid and Multilevel Preconditioners for Computational Photography


This paper unifies multigrid and multilevel (hierarchical) preconditioners, two widely-used approaches for solving computational photography and other computer graphic simulation problems. It also provides detailed experimental comparisons of these techniques and their variants, including a careful analysis of relative computational costs and how these impact practical algorithm performance.


Dilip Krishnan, New York University
Richard Szeliski, Microsoft Research