Subjective evaluations of example-based, total variation, and joint regularization for image processing

Anderson, Hyrum S.; Gupta, Manisha; Hardeberg, Jon Yngve

doi:10.1117/12.917710

Cited by 1 publication

(4 citation statements)

References 29 publications

(20 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Familiarity with the work of Anderson et al [22] made the study a good starting point to approach the solution of EB-SR. In this work, three different techniques are compared to result in the most efficient one regarding noisy deblurring (image SR) and denoising.…”

Section: Subjective Evaluations Of Eb Tv and Combined Regularizationmentioning

confidence: 84%

“…Although there are many approaches built for applying example-based super-resolution (EB-SR), the applicability of each one on the a specific dataset was questionable due to the morphology of the images. Thus, the three most promising state-of-the-art techniques [21,22,11] in literature were studied further and discussed with experts in the domain of SR, leading to the selection of two approaches [21,11] to be implemented, applied and evaluated.…”

Section: Image Quality Metricsmentioning

confidence: 99%

“…The combined regularisation proved to perform the best in the case of images with strong structure and when the dictionary is based on similar example images. The evaluations of the study were based on SSIM evaluation, which correlates best with human judgement of image quality [22].…”

Section: Subjective Evaluations Of Eb Tv and Combined Regularizationmentioning

confidence: 99%

“…For further technical procedures and mathematical background, the reader is directed to the documentation of the original study [22].…”

mentioning

confidence: 99%

See 3 more Smart Citations

Computer-Aided Reclamation of Lost Art

Δημητρίου¹,

Hardeberg²,

Adelmann³

2012

Computer Vision – ECCV 2012. Workshops and Demonstrations

View full text Add to dashboard Cite

Based on previous research on super-resolution and colour correction by example, the present project employs such techniques towards a high-quality reclamation of lost art. The aim is to produce high-quality images of nowadays destroyed or missing paintings using the correspondence between similar artworks available in both low and high quality images. Several approaches for both super-resolution and colour correction techniques have been studied, implemented and tested to result to the most efficient and appropriate to be used. This project is an attempt that has never been done before and the successful reclamation of lost art initialises a new area of colour imaging applications in fine art. It reveals unlimited possibilities in the domain and establishes the potential of further attempts in this direction.iii Computer-Aided Reclamation of Lost Art SummaryThe present document covers the entirety of a proposed approach and an implemented solution towards Reclamation of Lost Art. Employing a combination of example-based super-resolution and colour correction techniques, the project attempts a unique step in colour imaging applications in fine art. It explores the applicability and performance of state-of-the-art techniques in super-resolution, inverse halftoning, image registration, colour transfer and image blending methods to the demanding image set of fine art specimens.Firstly, in Chapter 1, the problem at hand is introduced and stated along with the motivation behind this aim. It expands on the significance of contribution of this endeavour, the implicated considerations and relevant previous research.

show abstract

Section: Subjective Evaluations Of Eb Tv and Combined Regularizationmentioning

confidence: 84%