Region filling and object removal by exemplar-based image inpainting

By Antonio Criminisi, Patrick Pérez, Kentaro Toyama. 2004

3,396 citations recorded by the source corpus.

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

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

Free-Form Image Inpainting With Gated Convolution (2018, methodology): “…2001; Barnes et al. 2009; Drori et al. 2003; Sun et al. 2005], structures [Huang et al. 2014], transformation or distortion information [Huang et al. 2013; Pavi´c et al. 2006] and image exemplars [Criminisi et al. 2004; Hays and Efros 2007; Kwatra et al. 2005; Whyte et al. 2009; Zhao et al. 2018].”
N4-Fields: Neural Network Nearest Neighbor Fields for Image Transforms (2014, methodology): “Most methods use nearest neighbor relations within the same image, e.g. Dabov et al. [17] for denoising or Criminisi et al. [18] for inpainting.”
Pluralistic Image Completion (2019, methodology): “Early image completion works [4, 7, 5, 8, 3, 13] focus only on steps 2 and 3 above, by assuming that gaps should be filled with similar content to that of the background.”
Incorporating Convolution Designs into Visual Transformers (2021, background): “For the locality, it is a common assumption in visual tasks [11, 26, 9] that neighboring pixels always tend to be correlated.”
Image Inpainting via Generative Multi-column Convolutional Neural Networks (2018, methodology): “Reliable Similar Patches In both exemplar-based [7, 8, 4, 21, 10, 11, 2] and recent learning-based methods [18, 25, 24, 9, 26], explicit nearest-neighbor search is one of the key components for generation of realistic details.”
The patch transform and its applications to image editing (2008, background): “Also related, in terms of goals and techniques, are the patch-based image synthesis methods [7, 9], which also require compatibility measures between patches.”
Articulated pose estimation with tiny synthetic videos (2015, methodology): “We estimate occluded pixel values using standard hole-filling algorithms [4, 6].”
Inpaint Anything: Segment Anything Meets Image Inpainting (2023, background): “Remove Anything focuses on the object removal problem [2, 3, 5] by allowing users to eliminate any object from an image while ensuring that the resulting image remains visually plausible.”
Image Inpainting by Patch Propagation Using Patch Sparsity (2010, methodology): “6(e), we also present the results of the most related algorithm in [22].”
Patch-Based Image Inpainting with Generative Adversarial Networks (2018, background): “Inpainting: Early inpainting studies, which worked on a single image, [2, 3, 22, 1] typically created solutions through filling the missing region with texture from similar or closest image areas, hence they suffered from the lack of global structural information.”
Text-Guided Neural Image Inpainting (2020, methodology): “Some methods are also based on patch match algorithms, but use external image sources [5, 43] such as search engine [34].”
Learning Oracle Attention for High-Fidelity Face Completion (2020, background): “[2] established the priority of each missing patch with reference to the surrounding structure.”
Learning Pyramid-Context Encoder Network for High-Quality Image Inpainting (2019, background): “High-quality image inpainting usually requires synthesizing not only visually-realistic but semanticallyreasonable content for missing regions [3, 5, 28, 29, 31].”
Local adaptivity to variable smoothness for exemplar-based image denoising and representation (2005, background): “In [25, 20 , 12, 13], several authors showed that the distance is a reliable measure to compare image patches.”
Sparse Representation for Color Image Restoration (2008, neutral): “et. al. for super-resolution [31], [32]; and by several follow-up works [ 33 ], [34], [35], [36] for”
Image Inpainting With Learnable Bidirectional Attention Maps (2019, background): “[4] suggested a gradient-based data term for filling in linear structure with higher priority.”
Recurrent Feature Reasoning for Image Inpainting (2020, background): “These algorithms have a wide range of applications in photo editing, de-captioning and other scenarios where people might want to remove unwanted objects from their photos [23, 22, 1, 4].”
Face De-Spoofing: Anti-Spoofing via Noise Modeling (2018, background): “In image inpainting, users mark the area to inpaint in a mask map and apply the filling based on the existing patch texture and the overall view structure in the unmasked region [39,36,37].”
Patch-based segmentation using expert priors: Application to hippocampus and ventricle segmentation (2011, methodology): “…patch-based methods are currently the focus of attention of the computer vision community in various domains such as texture synthesis (Efros and Freeman, 2001), in-painting (Criminisi et al., 2004), restoration (Buades et al., 2005), and single-frame super resolution (Protter et al., 2009).”