The patch transform and its applications to image editing

By Taeg Sang Cho, Moshe Butman, Shai Avidan, William T. Freeman. 2008 IEEE Conference on Computer Vision and Pattern Recognition, 2008

402 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/e452bd6870c883fc6da667c2450e8d39d0ab8152

How researchers see this paper

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

In Computer Vision and Pattern Recognition

Citations in the corpus

DoveNet: Deep Image Harmonization via Domain Verification (2019, methodology): “HAdobe5k: MIT-Adobe5k dataset [2] covers a wide range of scenes, objects, and lighting conditions.”
Image Reconstruction from Bag-of-Visual-Words (2014, background): “[50] solved the same problem as ours by introducing an approximation”
Learning to Restore Low-Light Images via Decomposition-and-Enhancement (2020, background): “Typical image enhancement methods [46, 51, 24, 7, 40, 34, 48, 4] propose to recover low-light images to match with human perception.”
Zero-Reference Deep Curve Estimation for Low-Light Image Enhancement (2020, methodology): “For example, the LLNet [16] was trained on data simulated on random Gamma correction; the LOL dataset [27] of paired low/normal light images was collected through altering the exposure time and ISO during image acquisition; the MIT-Adobe FiveK dataset [3] comprises 5,000 raw images, each of which ha”
A framework for photo-quality assessment and enhancement based on visual aesthetics (2010, methodology): “This is a significant improvement over a foreground object-centric image-editing technique [4], wherein the authors propose a method to reconstruct an image from low-resolution patches subject to user-defined constraints.”
Shift-map image editing (2009, methodology): “The method most related to our work is the patch transform [5], which segments the image into patches which are than rearranged using global optimization.”
Summarizing visual data using bidirectional similarity (2008, methodology): “Note that neither “Seam Carving”, nor “importance based summarization”, nor [5], exploit repetitiveness or redundancy of visual data in the retargeting/summarization process.”
SinGAN: Learning a Generative Model From a Single Natural Image (2019, background): “Classical examples include denoising [65], deblurring [39], super resolution [18], dehazing [2, 15], and image editing [37, 21, 9, 11, 50].”
Deep Image Harmonization (2017, methodology): “To cover more object categories, we augment it with the MIT-Adobe FiveK dataset [3].”
PatchMatch: a randomized correspondence algorithm for structural image editing (2009, background): “Image reshuffling algorithms make it possible to grab portions of the image and move them around – the computer automatically synthesizes the remainder of the image so as to resemble the original while respecting the moved regions [Simakov et al. 2008; Cho et al. 2008].”
Underexposed Photo Enhancement Using Deep Illumination Estimation (2019, methodology): “Figure 7: Visual comparison with state-of-the-art methods on a test image (a) from the MIT-Adobe FiveK [3] dataset.”
Fast Image Processing with Fully-Convolutional Networks (2017, methodology): “We use two image processing datasets: MIT-Adobe 5K and RAISE [12, 18].”
Image Inpainting : Overview and Recent Advances (2014, methodology): “The method, inspired from the patch transform described in [38], computes a vector field (called shift-map), which maps each pixel in the hole to be filled in to a pixel in the known part of the image.”
Learning to Enhance Low-Light Image via Zero-Reference Deep Curve Estimation (2021, methodology): “We did not use the MIT-Adobe FiveK dataset [24] as it is not primarily designed for underexposed photos enhancement and still contains some low-light images in the ground truth set.”