Contextual dictionaries for image super resolution

Authors:
Wei Yu;Hongxun Yao;Xianming Liu;Rongrong Ji;Xiaoshuai Sun;Pengfei Xu
Affiliations:
Harbin Institute of Technology;Harbin Institute of Technology;Harbin Institute of Technology;Harbin Institute of Technology;Harbin Institute of Technology;Harbin Institute of Technology
Venue:
Proceedings of the Third International Conference on Internet Multimedia Computing and Service
Year:
2011

Citing 4
Cited 0

Learning Low-Level Vision

International Journal of Computer Vision - Special issue on statistical and computational theories of vision: modeling, learning, sampling and computing, Part I
Limits on Super-Resolution and How to Break Them

IEEE Transactions on Pattern Analysis and Machine Intelligence
Double sparsity: learning sparse dictionaries for sparse signal approximation

IEEE Transactions on Signal Processing
Image Interpolation by Adaptive 2-D Autoregressive Modeling and Soft-Decision Estimation

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Traditional super resolution (SR) methods based on sparse representation have shown their excellent performance, however, the methods usually perform even worse when the input images and the training samples are diverse. Considering this problem, this paper presents a novel super resolution method based on sparse representation with contextual dictionary. Through adopting discriminative features instead of common features, the method train and use contextual dictionary in the SR process. Additionally, the method uses the first-order and second-order gradients of patch as representation, which ensures the neighbor information is introduced in the SR processing. The experiment results demonstrate the performance of this method has been promoted than other traditional method