Image Inpainting
276 papers with code • 12 benchmarks • 17 datasets
Image Inpainting is a task of reconstructing missing regions in an image. It is an important problem in computer vision and an essential functionality in many imaging and graphics applications, e.g. object removal, image restoration, manipulation, re-targeting, compositing, and image-based rendering.
Source: High-Resolution Image Inpainting with Iterative Confidence Feedback and Guided Upsampling
Image source: High-Resolution Image Inpainting with Iterative Confidence Feedback and Guided Upsampling
Benchmarks
These leaderboards are used to track progress in Image Inpainting
| Trend | Dataset | Best Model | Paper | Code | Compare |
|---|---|---|---|---|---|
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
|||||
|
Libraries
Use these libraries to find Image Inpainting models and implementations
ImageNet
CelebA
FFHQ
Places
CelebA-HQ
ApolloScape
Places365
Fashion-Gen
CASIA V2
FVI
Most implemented papers
Image Inpainting for Irregular Holes Using Partial Convolutions
NVIDIA/partialconv
•
•
ECCV 2018
Existing deep learning based image inpainting methods use a standard convolutional network over the corrupted image, using convolutional filter responses conditioned on both valid pixels as well as the substitute values in the masked holes (typically the mean value).
High-Resolution Image Synthesis with Latent Diffusion Models
compvis/stable-diffusion
•
•
CVPR 2022
By decomposing the image formation process into a sequential application of denoising autoencoders, diffusion models (DMs) achieve state-of-the-art synthesis results on image data and beyond.
Free-Form Image Inpainting with Gated Convolution
JiahuiYu/generative_inpainting
•
•
ICCV 2019
We present a generative image inpainting system to complete images with free-form mask and guidance.
Generative Image Inpainting with Contextual Attention
JiahuiYu/generative_inpainting
•
•
CVPR 2018
Motivated by these observations, we propose a new deep generative model-based approach which can not only synthesize novel image structures but also explicitly utilize surrounding image features as references during network training to make better predictions.
Implicit Neural Representations with Periodic Activation Functions
vsitzmann/siren
•
•
NeurIPS 2020
However, current network architectures for such implicit neural representations are incapable of modeling signals with fine detail, and fail to represent a signal's spatial and temporal derivatives, despite the fact that these are essential to many physical signals defined implicitly as the solution to partial differential equations.
EdgeConnect: Generative Image Inpainting with Adversarial Edge Learning
The edge generator hallucinates edges of the missing region (both regular and irregular) of the image, and the image completion network fills in the missing regions using hallucinated edges as a priori.
Deep Image Prior
DmitryUlyanov/deep-image-prior
•
•
CVPR 2018
In this paper, we show that, on the contrary, the structure of a generator network is sufficient to capture a great deal of low-level image statistics prior to any learning.
Generative Modeling by Estimating Gradients of the Data Distribution
ermongroup/ncsn
•
•
NeurIPS 2019
We introduce a new generative model where samples are produced via Langevin dynamics using gradients of the data distribution estimated with score matching.
Score-Based Generative Modeling through Stochastic Differential Equations
yang-song/score_sde
•
•
ICLR 2021
Combined with multiple architectural improvements, we achieve record-breaking performance for unconditional image generation on CIFAR-10 with an Inception score of 9. 89 and FID of 2. 20, a competitive likelihood of 2. 99 bits/dim, and demonstrate high fidelity generation of 1024 x 1024 images for the first time from a score-based generative model.
Consistency Models
Through extensive experiments, we demonstrate that they outperform existing distillation techniques for diffusion models in one- and few-step sampling, achieving the new state-of-the-art FID of 3. 55 on CIFAR-10 and 6. 20 on ImageNet 64x64 for one-step generation.
