Real-time texture synthesis by patch-based sampling

By Lin Liang, Ce Liu, Ying-Qing Xu, Baining Guo, Harry Shum. TOGS, 2001

769 citations recorded by the source corpus.

Subfields: Computational Mechanics, Computer Graphics and Computer-Aided Design, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/e02295601663f95d8df87adbe159cbd39b685543

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

In Computer Graphics and Computer-Aided Design

In Computer Vision and Pattern Recognition

Citations in the corpus

Motion texture: a two-level statistical model for character motion synthesis (2002, background): “Although the patches used in the patch-based texture synthesis may be regarded as textons as well, the concept of texton map was not explicitly discussed in [12, 25].”
Interactive image completion with perspective correction (2006, methodology): “Instead of this per-pixel procedure some approaches use larger patches or fragments [13, 25] in order to speed up the process.”
Fast Neighborhood Graph Search Using Cartesian Concatenation (2013, background): “It is also a crucial step in many vision and graphics problems, such as shape matching [13], object retrieval [29], feature matching [8, 35], texture synthesis [23], image completion [16] and so on.”
Demystifying Neural Style Transfer (2017, background): “The goal of style transfer is to generate a stylized image x ∗ given a content image x c and a reference style image x s .”
Aggregated Contextual Transformations for High-Resolution Image Inpainting (2021, methodology): “Inspired by patch-based methods for texture synthesis [31, 32], more and more patch-based filling approaches are proposed for image inpainting [3, 33].”
PatchMatch: a randomized correspondence algorithm for structural image editing (2009, methodology): “Further improvements modify the search and sampling approaches for better structure preservation [Wei and Levoy 2000; Ashikhmin 2001; Liang et al. 2001; Efros and Freeman 2001; Kwatra et al. 2003; Criminisi et al. 2003; Drori et al. 2003].”
Parallel controllable texture synthesis (2005, methodology): “Texture is created by iteratively overlapping irregular patches of the exemplar [Praun et al 2000] to minimize overlap error [Liang et al 2001].”
Image Completion Using Efficient Belief Propagation Via Priority Scheduling and Dynamic Pruning (2007, background): “An additional advantage we gain in this way is that after all MRF nodes have pruned their labels at least once (e.g., at the end of the first forward pass) then we can precompute the reduced matrices of pairwise potentials (which can now fit into memory) and, thus, greatly enhance the speed of our…”
Fast Patch-based Style Transfer of Arbitrary Style (2016, background): “While there is a large classical literature on texture synthesis methods that create artwork from a blank canvas [8, 20, 22, 35], several recent approaches study the problem of transferring the desired style from one image onto the structural content of another image.”
Image quilting for texture synthesis and transfer (2001, methodology): “[13] who report real-time performance using a very similar approach).”
Texture optimization for example-based synthesis (2005, neutral): “Among them, patchbased techniques [Efros and Freeman 2001; Liang et al. 2001; Cohen et al. 2003; Kwatra et al. 2003; Wu and Yu 2004] are generally more successful at synthesizing high quality textures as they can maintain global structure of the texture well.”
Object removal by exemplar-based inpainting (2003, methodology): “7c) and also reduces blocky artefacts without an expensive patch-cutting step [9] or a blur-inducing blending step [19].”
Nonparametric Scene Parsing via Label Transfer (2011, methodology): “In this section, we will describe each module of our nonparametric scene parsing system.”
Solid texture synthesis from 2D exemplars (2007, methodology): “First, we apply a PCA projection to the neighborhood vectors in the exemplar [Hertzmann et al. 2001; Liang et al. 2001; Lefebvre and Hoppe 2006].”
Image Inpainting : Overview and Recent Advances (2014, background): “Instead of synthesizing the missing region pixel per pixel, the idea of patch-based solutions is to recover entire patches in one step by sampling and copying texture patterns (entire patches) from the source [20].”
A Learned Representation For Artistic Style (2016, positive): “Efros & Freeman (2001) and Liang et al. (2001) extend this idea to “growing” textures one patch at a time, and Efros & Freeman (2001) uses the approach to implement “texture transfer”, i.e. transfering the texture of an object onto another one.”
A two-step approach to hallucinating faces: global parametric model and local nonparametric model (2001, methodology): “Since patch-based nonparametric sampling has been applied in texture synthesis with encouraging results and high efficiency [8, 13, 5], we build a patch-based Markov network, a nonparametric counterpart to the parametric one proposed in [6], to model the statistics between a local feature image and ”
Face Hallucination: Theory and Practice (2007, background): “(18), (20) and (22), we obtain X∗ = (I − Q2(QT 2 Λ−1Q2)−1QT 2 Λ−1)”