Bayesian Super-Resolution of Text in Video with a Text-Specific Bimodal Prior

Donaldson, K.; Myers, Gregory

doi:10.1109/cvpr.2005.87

Cited by 30 publications

(35 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The above affine photometric model can only handle small photometric changes, therefore it has been extended to a non-linear model in [289], which also [113], [115], [116], [122], [128], [133], [138], [147], [149], [170], [174], [177], [181], [186], [187], [188], [189], [199], [204], [209], [215], [216], [217], [218], [223], [224], [226], [227], [228], [230], [231], [235], [242], [246], [248], [249], [250], [251], [252], [258], [260], [270], [272], [276], [277], [279], [280], …”

Section: Imaging Modelsmentioning

confidence: 99%

Super-resolution: a comprehensive survey

Nasrollahi

Moeslund

2014

Machine Vision and Applications

586

292

View full text Add to dashboard Cite

Super-resolution, the process of obtaining one or more high-resolution images from one or more low-resolution observations, has been a very attractive research topic over the last two decades. It has found practical applications in many real world problems in different fields, from satellite and aerial imaging to medical image processing, to facial image analysis, text image analysis, sign and number plates reading, and biometrics recognition, to name a few. This has resulted in many research papers, each developing a new superresolution algorithm for a specific purpose. The current comprehensive survey provides an overview of most of these published works by grouping them in a broad taxonomy. For each of the groups in the taxonomy, the basic concepts of the algorithms are first explained and then the paths through which each of these groups have evolved are given in detail, by mentioning the contributions of different authors to the basic concepts of each group. Furthermore, common issues in super-resolution algorithms, such as imaging models and registration algorithms, optimization of the cost functions employed, dealing with color information, improvement factors, assessment of super-resolution algorithms, and the most commonly employed databases are discussed.

show abstract

Section: Imaging Modelsmentioning

confidence: 99%

Super-resolution: a comprehensive survey

Nasrollahi

Moeslund

2014

Machine Vision and Applications

586

292

View full text Add to dashboard Cite

show abstract

“…Bayesian estimation and probabilistic graphical models are a powerful tool which allow to model the statistical distributions of observed and hidden values and to take into account the statistical relationships between them. Although there is a wide variety in model families and types, in the domain of document segmentation the most frequently used models are flat Markov random fields (MRF) [1,2,3,4,5].…”

Section: Introductionmentioning

confidence: 99%

Families of Markov models for document image segmentation

Wolf

2009

2009 IEEE International Workshop on Machine Learning for Signal Processing

View full text Add to dashboard Cite

In this paper we compare several directed and undirected graphical models for different image segmentation problems in the domain of document image processing and analysis. We show that adapting the structure of the model to specific sitations at hand, for instance character restoration, recto/verso separation and segmenting high resolution character images, can significantly improve segmentation performance. We propose inference algorithms for the different models and we test them on different data sets (manuscripts and printed text of different qualities).

show abstract

“…This encourages local smoothness while preserving any step edge sharpness. Donaldson and Myers [23] used the same Huber gradient penalty function with an additional prior probability distribution based on the bimodal characteristic of text. The MAP estimator with the Huber penalty prior term provides slightly smoother results.…”

Section: Maximum a Posteriori Estimator (Map) -mentioning

confidence: 99%

“…In the spatial domain, Donaldson and Myers [23] used pairwise correlation over the whole image with quadratic interpolation and a least-squares fit to determine the translation vector for each observed LR frame. Li and Doermann [74] performed sub-pixel registration by first bilinearly interpolating frames and then by using correlation minimizing Sum of Square Difference (SSD) between text blocks.…”

Section: Context Of Super-resolution Algorithmsmentioning

confidence: 99%

“…Donaldson and Myers [23] also assumed a pure translational model and motion estimation was carried out by pairwise correlation. A Bayesian framework with a maximum a posteriori (MAP) estimator was then used for reconstruction of SR text which allowed the inclusion of a priori information to constrain errors: a bimodality prior assuming that text is bimodal and a Gibbs prior with a Huber gradient penalty function assuming that text images are locally smooth.…”

Section: Pre-processing Steps Of Text Extractionmentioning

confidence: 99%

See 1 more Smart Citation