Single Image Super-Resolution

Image super-resolution is the task of obtaining a high-resolution (HR) image of a scene given low-resolution (LR) image(s) of the scene. In this project, we have focused on the task of super-resolution given a single LR image, which is usually the case. There exist many techniques in literature addressing this task, and we have considered two techniques having the essence of [1] and [2]. In first approach, HR patches are constructed using a sparse representation of the corresponding LR patch in a compact dictionary [1]. In the other approach, we exploit the repetition of local visual content within and across different scales of the given LR image[2].

[1]  Sunil Arya,et al.  Approximate nearest neighbor queries in fixed dimensions , 1993, SODA '93.

[2]  William T. Freeman,et al.  Learning low-level vision , 1999, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[3]  Takeo Kanade,et al.  Limits on super-resolution and how to break them , 2000, Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662).

[4]  Takeo Kanade,et al.  Hallucinating faces , 2000, Proceedings Fourth IEEE International Conference on Automatic Face and Gesture Recognition (Cat. No. PR00580).

[5]  Takeo Kanade,et al.  Limits on Super-Resolution and How to Break Them , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[6]  William T. Freeman,et al.  Example-Based Super-Resolution , 2002, IEEE Computer Graphics and Applications.

[7]  Moon Gi Kang,et al.  Super-resolution image reconstruction: a technical overview , 2003, IEEE Signal Process. Mag..

[8]  Michal Irani,et al.  Super-resolution from a single image , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[9]  Thomas S. Huang,et al.  Image Super-Resolution Via Sparse Representation , 2010, IEEE Transactions on Image Processing.