Graphcut textures: image and video synthesis using graph cuts

By Vivek Kwatra, Arno Schödl, Irfan Essa, Greg Turk, Aaron F. Bobick. ACM Transactions on Graphics, 2003

1,725 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/55c3ecb47aebba55a9de761c0e4dc0fd6d8d28b0

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

Photorealistic Facial Texture Inference Using Deep Neural Networks (2016, methodology): “Non-parametric synthesis algorithms [14, 51, 13, 26] have been developed to synthesize repeating structures using samples from small patches, while ensuring local consistency.”
Image Inpainting : Overview and Recent Advances (2014, background): “Graph cuts have been used for a wide variety of vision problems, particularly in [44] to seamlessly combine and stitch different textures together.”
Image Completion Using Global Optimization (2006, background): “Here one tries to ﬁll the unknown region simply by copying pixels or source patches from the observed part of the image [9, 4].”
Space-Time Completion of Video (2007, methodology): “The two drawbacks of this approach are that elaborate methods for combining the large patches are needed for hiding inconsistencies [12], [13], [21], and that the data set needs to be artificially enlarged by including various skewed and scaled replicas that might be needed for completion.”
Image Completion Using Efficient Belief Propagation Via Priority Scheduling and Dynamic Pruning (2007, neutral): “13(a).”
Two-Stream Convolutional Networks for Dynamic Texture Synthesis (2017, background): “There are two general approaches that have dominated the texture synthesis literature: non-parametric sampling approaches that synthesize a texture by sampling pixels of a given source texture [10, 26, 37, 47], and statistical parametric models.”
Get Out of my Picture! Internet-based Inpainting (2009, methodology): “The problem of how best to combine two images whose properties do not necessarily match has been approached in many ways, from methods which aim to conceal boundaries between regions, such as Burt & Adelson’s multiresolution spline [9] and Poisson blending [22], to methods that try and find the best”
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].”
Combining Markov Random Fields and Convolutional Neural Networks for Image Synthesis (2016, methodology): “This opens an interesting future work that how the dCNN can be retrained, or incorporated with stitching based texture synthesis such as [15] for achieving pixel-level photorealism.”
Higher Order SVD Analysis for Dynamic Texture Synthesis (2008, background): “This type of synthesis is called nonparametric or patch-based, since it is not based on a model and reduces the synthesis to a collage of patches.”
Kernel Learning for Dynamic Texture Synthesis (2016, background): “As previously mentioned, almost all the current dynamic system based DT synthesis remain two main issues: 1) their stabilities are very hard to guarantee; and 2) they require the full set of training video frames.”
Neural Architecture Search for Deep Image Prior (2020, background): “Several in-painting works exploit MRF formulations of this kind [22, 18, 24], including methods that source patches from the input [10] or multiple external [17, 14] images, or use random propagation of a few good matched patches [3].”