A wavelet-based multiresolution statistical model for texture

Zhang, J.; Wang, D.; Tran, Quang-Anh

doi:10.1109/83.725370

Cited by 26 publications

(9 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The gradient of holds information about the phase and direction changes of the structure. Obtaining the gradient of with respect to yields (8) where . Then (9) where , and .…”

Section: ) Structural Componentmentioning

confidence: 99%

See 1 more Smart Citation

Texture synthesis: textons revisited

Charalampidis

2006

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

This paper introduces a technique for synthesizing natural textures, with emphasis on quasiperiodic and structural textures. Textures are assumed to be composed of three components, namely illumination, structure, and stochastic. The contribution of this work is that, in contrast to previous techniques, it proposes a joint approach for handling the texture's global illumination, irregular structure, and stochastic component which may be correlated to the other two components. Furthermore, the proposed technique does not produce verbatim copies in the synthesized texture. More specifically, a top-down approach is used for extraction of texture elements (textons) in which, in contrast to previous texton-based approaches, no assumptions regarding perfect periodicity are made. The structure itself can be modeled as a stochastic process. Consequently, textons are allowed to have irregular and nonidentical shapes. In the synthesis stage, a new nonregular textural structure is designed from the original one that defines the place holders for textons. We call such place holders empty textons (e-textons). The e-textons are filled in by a representative texton. Since e-textons do not have identical shapes, a texton shape-matching procedure is required. After adding the illumination to the structural component, a strictly localized version of a block sampling technique is applied to add the stochastic component. The block sampling technique combined with the addition of the illumination component provides a significant improvement in the appearance of synthesized textures. Results show that the proposed method is successful in synthesizing structural textures visually indistinguishable to the original. Moreover, the method is successful in synthesizing a variety of stochastic textures.

show abstract

“…The gradient of holds information about the phase and direction changes of the structure. Obtaining the gradient of with respect to yields (8) where . Then (9) where , and .…”

Section: ) Structural Componentmentioning

confidence: 99%

“…Multiresolution approaches are common in texture synthesis. Zhang et al [8] proposed a multiresolution statistical model, consisting of random fields in wavelet sub-bands. A complex random field model containing long-range and nonlinear spatial correlations is constructed from several simpler ones.…”

Section: Introductionmentioning

confidence: 99%

Texture synthesis: textons revisited

Charalampidis

2006

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

show abstract

“…building up a low resolution approximation to the texture and then refining it to higher resolutions) and fast search algorithms [WLOO]. Performing the texture synthesis in the wavelet domain has allowed proper separation of information at different spatial scales [ZWT98], including the very efficient approximation using a neighbourhood consisting of only lower resolution subbands [DEB97].…”

Section: Literature Reviewmentioning

confidence: 99%

Towards Realism in Facial Image Transformation: Results of a Wavelet MRF Method

Tiddeman

Stirrat

Perrett

2005

Computer Graphics Forum

View full text Add to dashboard Cite

The ability to transform facial images between groups (e.g. from young to old, or from male to female) has applications in psychological research, police investigations, medicine and entertainment. Current techniques suffer either from a lack of realism due to unrealistic or inappropriate textures in the output images, or a lack of statistical validity, e.g. by using only a single example image for training. This paper describes a new method for improving the realism and effectiveness of facial transformations (e.g. ageing, feminising etc.) of individuals. The method aims to transform low resolution image data using the mean differences between the two groups, but converges on more specific texture features at the finer resolutions. We separate high and low resolution information by transforming the image into a wavelet domain. At each point we calculate a mapping from the original set to the target set based on the probability distributions of the input and output wavelet values. These distributions are estimated from the example images, using the assumption that the distribution depends on the values in a local neighbourhood of the point (the Markov Random Field (MRF) assumption). We use a causal neighbourhood that spans multiple coarser scales of the wavelet pyramid. The distributions are estimated by smoothing the histogram of example values. By increasing the smoothing of the histograms at coarser resolutions we are able to maintain perceived identity across the transforms while producing realistic fine-scale textures. We use perceptual testing to validate the new method, and the results show that it can produce more accurate shifts in perceived age and an increase in realism.

show abstract

“…32,33 This initial proposal has been followed by several studies on texture analysis using wavelet transform. [34][35][36][37][38][39][40][41][42][43] The texture classification rate of the conventional pyramid structured wavelet transform is higher than that of other methods such as Gabor filter and Tree structured wavelet transform, when more number of features is used. 34 The concept of Multi Resolution Markov Random Field (MRMRF) modeling is proposed in Ref.…”

Section: Introductionmentioning

confidence: 99%

Texture Characterization Using WSFS and WCFS

Arivazhagan¹,

Ganesan

2009

Int. J. Image Grap.

View full text Add to dashboard Cite

Texture analysis plays an important role in many tasks, ranging from remote sensing to medical imaging and query by content in large image data bases. The main difficulty of texture analysis in the past was the lack of adequate tools to characterize different scales of textures effectively. The development in multi-resolution analysis such as Gabor and wavelet transform help to overcome this difficulty. This paper describes the texture characterization system using (i) Wavelet Statistical Features (WSF) obtained from 3-level Discrete Wavelet Transformed (DWT) images, (ii) Wavelet Co-occurrence Features (WCF) of one level wavelet transformed images in detail.

show abstract

A wavelet-based multiresolution statistical model for texture

Cited by 26 publications

References 13 publications

Texture synthesis: textons revisited

Texture synthesis: textons revisited

Towards Realism in Facial Image Transformation: Results of a Wavelet MRF Method

Texture Characterization Using WSFS and WCFS

Contact Info

Product

Resources

About