Image inpainting
By Marcelo Bertalmío, Guillermo Sapiro, V. Caselles, Coloma Ballester. 2000
4,400 citations recorded by the source corpus.
Inpainting, the technique of modifying an image in an undetectable form, is as ancient as art itself. The goals and applications of inpainting are numerous, from the restoration of damaged paintings and photographs to the removal/replacement of selected objects. In this paper, we introduce a novel algorithm for digital inpainting of still images that attempts to replicate the basic techniques used by professional restorators. After the user selects the regions to be restored, the algorithm automatically fills-in these regions with information surrounding them. The fill-in is done in such a way that isophote lines arriving at the regions' boundaries are completed inside. In contrast with previous approaches, the technique here introduced does not require the user to specify where the novel information comes from. This is automatically done (and in a fast way), thereby allowing to simultaneously fill-in numerous regions containing completely different structures and surrounding backgrounds. In addition, no limitations are imposed on the topology of the region to be inpainted. Applications of this technique include the restoration of old photographs and damaged film; removal of superimposed text like dates, subtitles, or publicity; and the removal of entire objects from the image like microphones or wires in special effects.
Subfields: 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 Computer Graphics and Computer-Aided Design
- Jia-Bin Huang
- Eli Shechtman
- Kaiming He
- Johannes Peter Kopf
- Jonathan T. Barron
- Ce Liu
- Daniel Cohen-Or
- William T. Freeman
In Computer Vision and Pattern Recognition
- Luc Van Gool
- Ming-Hsuan Yang
- Dacheng Tao
- Xiaogang Wang
- Trevor Darrell
- Jitendra Malik
- Andrew Zisserman
- Andrea Vedaldi
Citations in the corpus
- Context Encoders: Feature Learning by Inpainting (2016, background): “Inpainting and hole-filling It is important to point out that our hole-filling task cannot be handled by classical inpainting [4, 32] or texture synthesis [2, 11] approaches, since the missing region is too large for local non-semantic methods to work well.”
- EdgeConnect: Generative Image Inpainting with Adversarial Edge Learning (2019, methodology): “Diffusion-based methods propagate neighboring information into the missing regions [4, 2].”
- Free-Form Image Inpainting With Gated Convolution (2018, background): “ly create novel objects in images. 2 RELATED WORK 2.1 Automatic Image Inpainting A variety of approaches have been proposed for image inpainting. Traditionally, di‡usion-based [Ballester et al. 2001; Bertalmio et al. 2000] methods propagate local image appearance next to the target holes based on th”
- Semantic Image Inpainting with Deep Generative Models (2016, background): “Numerous applications such as restoration of damaged paintings or image editing [3] benefit from accurate semantic inpainting methods if large regions are missing.”
- Palette: Image-to-Image Diffusion Models (2021, background): “Early inpainting approaches [Barnes et al. 2009; Bertalmio et al. 2000; Hays and Efros 2007; He and Sun 2012] work well on textured regions but often fall short in generating semantically consistent structure.”
- Generative Image Inpainting with Contextual Attention (2018, background): “Traditional diffusion or patch-based approaches such as [2, 4, 9, 10] typically use variational algorithms or patch similarity to propagate information from the background regions to the holes.”
- Sum-product networks: A new deep architecture (2011, background): “, [31, 3]), but the focus tends to be restoring smal l occlusions (e.”
- Image Inpainting for Irregular Holes Using Partial Convolutions (2018, background): “Non-learning approaches to image inpainting rely on propagating appearance information from neighboring pixels to the target region using some mechanisms like distance field[3, 1, 30].”
- Self-Supervised Feature Learning by Learning to Spot Artifacts (2018, background): “A first approach to create artifacts is to use inpainting algorithms [2, 6].”
- GRAND: Graph Neural Diffusion (2021, methodology): “, 1997; Chan & Vese, 2001) and inpainting (Bertalmio et al., 2000).”
- Variable Exponent, Linear Growth Functionals in Image Restoration (2006, background): “6) also has direct application in image processing, as it can be used for image interpolation [13, 27], also referred to as noise-free image inpainting [8, 17, 18].”
- Variational PDE models in image processing (2002, background): “Other approaches to the inpainting problem can be found in [3, 5, 8].”
- Inpainting of Binary Images Using the Cahn–Hilliard Equation (2007, background): “The pioneering work of Bertalmio et al. [1] introduced image inpainting for digital image processing.”
- Globally and locally consistent image completion (2017, methodology): “For example, the propagation can be performed based on the isophote direction field [Ballester et al. 2001; Bertalmio et al. 2000], or global image statistics based on the histograms of local features [Levin et al.”
- Efficient Tensor Completion for Color Image and Video Recovery: Low-Rank Tensor Train (2016, background): “The common target in LRTC is to recover missing entries of a tensor from its partially observed entities [22]–[24].”
- Eye In-painting with Exemplar Generative Adversarial Networks (2017, background): “In-painting has been studied extensively, both with and without deep networks [5, 38, 29].”
- Image Inpainting by End-to-End Cascaded Refinement With Mask Awareness (2021, methodology): “Partial differential equation (PDE) based methods [24]–[26] are designed to connect edges or to extend lines”