FRAME: filters, random fields, and minimax entropy towards a unified theory for texture modeling

Zhu, Song Chun; Wu, Yingnian; Mumford, David

doi:10.1109/cvpr.1996.517147

Cited by 43 publications

(45 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…More recently, De Bonet [4] and Efros and Leung [11] showed that a nearest-neighbor search can perform high-quality texture synthesis in a single pass, using multiscale and single-scale neighborhoods, respectively. (This search may be viewed as an approximation to sampling from an MRF, an approach used by Zhu et al [54] and Portilla and Simoncelli [40].) Wei and Levoy [49] unify these approaches, using neighborhoods consisting of pixels both at the same scale and at coarser scales.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Image analogies

Hertzmann

Jacobs

Oliver

et al. 2001

Proceedings of the 28th Annual Conference on Computer Graphics and Interactive Techniques

1,410

995

View full text Add to dashboard Cite

Figure 1 An image analogy. Our problem is to compute a new "analogous" image B that relates to B in "the same way" as A relates to A. Here, A, A , and B are inputs to our algorithm, and B is the output. The full-size images are shown in Figures 10 and 11. AbstractThis paper describes a new framework for processing images by example, called "image analogies." The framework involves two stages: a design phase, in which a pair of images, with one image purported to be a "filtered" version of the other, is presented as "training data"; and an application phase, in which the learned filter is applied to some new target image in order to create an "analogous" filtered result. Image analogies are based on a simple multiscale autoregression, inspired primarily by recent results in texture synthesis. By choosing different types of source image pairs as input, the framework supports a wide variety of "image filter" effects, including traditional image filters, such as blurring or embossing; improved texture synthesis, in which some textures are synthesized with higher quality than by previous approaches; super-resolution, in which a higher-resolution image is inferred from a low-resolution source; texture transfer, in which images are "texturized" with some arbitrary source texture; artistic filters, in which various drawing and painting styles are synthesized based on scanned real-world examples; and texture-by-numbers, in which realistic scenes, composed of a variety of textures, are created using a simple painting interface.Please see http://grail.cs.washington.edu/projects/image-analogies/ for additional information and results.While image analogies are clearly a desirable goal, it is not so clear how they might be achieved.

show abstract

Section: Related Workmentioning

confidence: 99%

“…Another way to improve the perceptual results of matching is to compute multiple scales of oriented derivative filters [4,23,54]. To this end, we can compute a steerable pyramid [45] for the luminance of A and B and concatenate the filter responses to the feature vectors for these images.…”

Section: Featuresmentioning

confidence: 99%

Image analogies

Hertzmann

Jacobs

Oliver

et al. 2001

Proceedings of the 28th Annual Conference on Computer Graphics and Interactive Techniques

1,410

995

View full text Add to dashboard Cite

show abstract

“…Both these results are consistent with the smooth nature of the samples that are generated by a FoE trained on natural images, supporting the idea that the FoE primarily models smoothness constraints with a robust loss function. Furthermore, Zhu and Mumford [17] and Zhu et al [18] have proposed a model which is based on fixed filters but the potential function is non-parametric, so that there are no restrictions on the shape of the learned potentials. They find that decaying potentials (such as the Student-t potential) are not sufficient in order to obtain good generative models, which raises the question of how the choice of a particular parametric form for the potential function (such as the Student-t potential) affects the expressiveness of the FoE.…”

Section: Field Of Expertsmentioning

confidence: 99%

Learning generative texture models with extended Fields-of-Experts

Heess¹,

Williams²,

Hinton

2009

Procedings of the British Machine Vision Conference 2009

View full text Add to dashboard Cite

We evaluate the ability of the popular Field-of-Experts (FoE) to model structure in images. As a test case we focus on modeling synthetic and natural textures. We find that even for modeling single textures, the FoE provides insufficient flexibility to learn good generative models -it does not perform any better than the much simpler Gaussian FoE. We propose an extended version of the FoE (allowing for bimodal potentials) and demonstrate that this novel formulation, when trained with a better approximation of the likelihood gradient, gives rise to a more powerful generative model of specific visual structure that produces significantly better results for the texture task.

show abstract

“…Our work is perhaps most fundamentally motivated by the impressive advances in texture synthesis methods [17,7,32,44,9,39] which have made it possible to create, for an increasingly wide range of patterns, unlimited quantities of a texture that is perceptually equivalent to a small provided sample.…”

Section: Previous Workmentioning

confidence: 99%

“…This method yields impressive results for an even wider variety of patterns, though some difficulties remain in preserving larger scale globally significant structure. Several other highly sophisticated texture analysis/synthesis approaches have been subsequently developed [9,32,44]. Of these, we chose to follow the texture synthesis approach proposed by Efros and Leung [9] because of its combination of simplicity and quality of output.…”

Section: Previous Workmentioning

confidence: 99%

Texture synthesis for 3d shape representation

Gorla¹,

Interrante

Sapiro

2003

IEEE Trans. Visual. Comput. Graphics

View full text Add to dashboard Cite

Considerable evidence suggests that a viewer's perception of the 3D shape of a polygonally-defined object can be significantly affected (either masked or enhanced) by the presence of a surface texture pattern. However investigations into the specific mechanisms of texture's effect on shape perception are still ongoing and the question of how to design and apply a texture pattern to a surface in order to best facilitate shape perception remains open. Recently, we have suggested that for anisotropic texture patterns, the accuracy of shape judgments may be significantly affected by the orientation of the surface texture pattern anisotropy with respect to the principal directions of curvature over the surface. However it has been difficult, until this time, to conduct controlled studies specifically investigating the effect of texture orientation on shape perception because there has been no simple and reliable method for texturing an arbitrary doubly curved surface with a specified input pattern such that the dominant orientation of the pattern everywhere follows a pre-defined directional vector field over the surface, while seams and projective distortion of the pattern are avoided. In this paper, we present a straightforward and highly efficient method for achieving such a texture and describe how it can potentially be used to enhance shape representation. Specifically, we describe a novel, efficient, automatic algorithm for seamlessly synthesizing, from a sample 2D pattern, a high resolution fitted surface texture in which the dominant orientation of the pattern locally follows a specified vector field over the surface at a per-pixel level, and in which seams, projective distortion, and repetition artifacts in the texture pattern are nearly completely avoided. We demonstrate the robustness of our method with a variety of texture swatches applied to standard graphics datasets, and we explain how our method can be used to facilitate research in the perception of shape from texture.

show abstract

FRAME: filters, random fields, and minimax entropy towards a unified theory for texture modeling

Cited by 43 publications

References 10 publications

Image analogies

Image analogies

Learning generative texture models with extended Fields-of-Experts

Texture synthesis for 3d shape representation

Contact Info

Product

Resources

About