Algorithms for super-resolution of images and videos based on learning methods

With super-resolution (SR) we refer to a class of techniques that enhance the spatial resolution of images and videos. SR algorithms can be of two kinds: multi-frame methods, where multiple low-resolution images are aggregated to form a unique high-resolution image, and single-image methods, that ai...

Full description

Bibliographic Details
Main Author: Bevilacqua, Marco
Language:English
Published: Université Rennes 1 2014
Subjects:
Online Access:http://tel.archives-ouvertes.fr/tel-01064396
http://tel.archives-ouvertes.fr/docs/01/06/43/96/PDF/BEVILACQUA_Marco.pdf
id ndltd-CCSD-oai-tel.archives-ouvertes.fr-tel-01064396
record_format oai_dc
spelling ndltd-CCSD-oai-tel.archives-ouvertes.fr-tel-010643962014-09-17T03:26:58Z http://tel.archives-ouvertes.fr/tel-01064396 2014REN1S027 http://tel.archives-ouvertes.fr/docs/01/06/43/96/PDF/BEVILACQUA_Marco.pdf Algorithms for super-resolution of images and videos based on learning methods Bevilacqua, Marco [INFO:INFO_TI] Computer Science/Image Processing [INFO:INFO_TI] Informatique/Traitement des images Algorithms Image processing Machine learning With super-resolution (SR) we refer to a class of techniques that enhance the spatial resolution of images and videos. SR algorithms can be of two kinds: multi-frame methods, where multiple low-resolution images are aggregated to form a unique high-resolution image, and single-image methods, that aim at upscaling a single image. This thesis focuses on developing theory and algorithms for the single-image SR problem. In particular, we adopt the so called example-based approach, where the output image is estimated with machine learning techniques, by using the information contained in a dictionary of image "examples". The examples consist in image patches, which are either extracted from external images or derived from the input image itself. For both kinds of dictionary, we design novel SR algorithms, with new upscaling and dictionary construction procedures, and compare them to state-of-the-art methods. The results achieved are shown to be very competitive both in terms of visual quality of the super-resolved images and computational complexity. We then apply our designed algorithms to the video upscaling case, where the goal is to enlarge the resolution of an entire video sequence. The algorithms, opportunely adapted to deal with this case, are also analyzed in the coding context. The analysis conducted shows that, in specific cases, SR can also be an effective tool for video compression, thus opening new interesting perspectives. 2014-06-04 eng PhD thesis Université Rennes 1
collection NDLTD
language English
sources NDLTD
topic [INFO:INFO_TI] Computer Science/Image Processing
[INFO:INFO_TI] Informatique/Traitement des images
Algorithms
Image processing
Machine learning
spellingShingle [INFO:INFO_TI] Computer Science/Image Processing
[INFO:INFO_TI] Informatique/Traitement des images
Algorithms
Image processing
Machine learning
Bevilacqua, Marco
Algorithms for super-resolution of images and videos based on learning methods
description With super-resolution (SR) we refer to a class of techniques that enhance the spatial resolution of images and videos. SR algorithms can be of two kinds: multi-frame methods, where multiple low-resolution images are aggregated to form a unique high-resolution image, and single-image methods, that aim at upscaling a single image. This thesis focuses on developing theory and algorithms for the single-image SR problem. In particular, we adopt the so called example-based approach, where the output image is estimated with machine learning techniques, by using the information contained in a dictionary of image "examples". The examples consist in image patches, which are either extracted from external images or derived from the input image itself. For both kinds of dictionary, we design novel SR algorithms, with new upscaling and dictionary construction procedures, and compare them to state-of-the-art methods. The results achieved are shown to be very competitive both in terms of visual quality of the super-resolved images and computational complexity. We then apply our designed algorithms to the video upscaling case, where the goal is to enlarge the resolution of an entire video sequence. The algorithms, opportunely adapted to deal with this case, are also analyzed in the coding context. The analysis conducted shows that, in specific cases, SR can also be an effective tool for video compression, thus opening new interesting perspectives.
author Bevilacqua, Marco
author_facet Bevilacqua, Marco
author_sort Bevilacqua, Marco
title Algorithms for super-resolution of images and videos based on learning methods
title_short Algorithms for super-resolution of images and videos based on learning methods
title_full Algorithms for super-resolution of images and videos based on learning methods
title_fullStr Algorithms for super-resolution of images and videos based on learning methods
title_full_unstemmed Algorithms for super-resolution of images and videos based on learning methods
title_sort algorithms for super-resolution of images and videos based on learning methods
publisher Université Rennes 1
publishDate 2014
url http://tel.archives-ouvertes.fr/tel-01064396
http://tel.archives-ouvertes.fr/docs/01/06/43/96/PDF/BEVILACQUA_Marco.pdf
work_keys_str_mv AT bevilacquamarco algorithmsforsuperresolutionofimagesandvideosbasedonlearningmethods
_version_ 1716714191374516224