Self-content super-resolution for ultra-HD up-sampling

Authors:
Mehmet Türkan;Dominique Thoreau;Philippe Guillotel
Affiliations:
Technicolor R&D France, Cesson Séévigné, France;Technicolor R&D France, Cesson Séévigné, France;Technicolor R&D France, Cesson Séévigné, France
Venue:
Proceedings of the 9th European Conference on Visual Media Production
Year:
2012

Citing 21
Cited 0

Improving resolution by image registration

CVGIP: Graphical Models and Image Processing
Fast texture synthesis using tree-structured vector quantization

Proceedings of the 27th annual conference on Computer graphics and interactive techniques
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
Synthesizing natural textures

I3D '01 Proceedings of the 2001 symposium on Interactive 3D graphics
Image analogies

Proceedings of the 28th annual conference on Computer graphics and interactive techniques
Example-Based Super-Resolution

IEEE Computer Graphics and Applications
Texture Synthesis by Non-Parametric Sampling

ICCV '99 Proceedings of the International Conference on Computer Vision-Volume 2 - Volume 2
Fundamental Limits of Reconstruction-Based Superresolution Algorithms under Local Translation

IEEE Transactions on Pattern Analysis and Machine Intelligence
A Non-Local Algorithm for Image Denoising

CVPR '05 Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Volume 2 - Volume 02
Image upsampling via imposed edge statistics

ACM SIGGRAPH 2007 papers
Example-Based Learning for Single-Image Super-Resolution

Proceedings of the 30th DAGM symposium on Pattern Recognition
Image super-resolution via sparse representation

IEEE Transactions on Image Processing
Efficient Parallel Algorithm for Nonlinear Dimensionality Reduction on GPU

GRC '10 Proceedings of the 2010 IEEE International Conference on Granular Computing
Image and video upscaling from local self-examples

ACM Transactions on Graphics (TOG)
Matching pursuits with time-frequency dictionaries

IEEE Transactions on Signal Processing
A computationally efficient superresolution image reconstruction algorithm

IEEE Transactions on Image Processing
Image quality assessment: from error visibility to structural similarity

IEEE Transactions on Image Processing
Region filling and object removal by exemplar-based image inpainting

IEEE Transactions on Image Processing
Fast and robust multiframe super resolution

IEEE Transactions on Image Processing
Nonlinear approximation based image recovery using adaptive sparse reconstructions and iterated denoising-part I: theory

IEEE Transactions on Image Processing
Nonlinear approximation based image recovery using adaptive sparse reconstructions and iterated denoising-part II: adaptive algorithms

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

We describe a self-content single image super-resolution algorithm based on multi-scale neighbor embeddings of image patches. We make use of the recurrence property of similar patches across different scales of an image. Inspired by manifold learning approaches, we first characterize the local geometry of a given low-resolution patch by reconstructing it from similar patches taken from down-scaled versions of the input image. We then hallucinate the high-resolution patch by relying on local geometric similarities of low- and high-resolution patch spaces. We enforce the local compatibility through patch overlapping, and preserve the structures with a sparsity-based patch averaging. We further enforce the global consistency through back-projection. Noting that this method uses as little as the self-information contained in a given low-resolution image and its implementation is well suited to GPU processors thanks to highly parallelization, our experimental results demonstrate similar or even better performance with respect to state-of-the-art superresolution methods.