mlsanityData to 7 Oct 2026

Generating natural images with direct Patch Distributions Matching

arXiv 2203.11862 ↗First posted 22 Mar 2022cs.CVcited by 26

Not on the map. Why?

Abstract (arXiv)

Many traditional computer vision algorithms generate realistic images by requiring that each patch in the generated image be similar to a patch in a training image and vice versa. Recently, this classical approach has been replaced by adversarial training with a patch discriminator. The adversarial approach avoids the computational burden of finding nearest neighbors of patches but often requires very long training times and may fail to match the distribution of patches. In this paper we leverage the recently developed Sliced Wasserstein Distance and develop an algorithm that explicitly and efficiently minimizes the distance between patch distributions in two images. Our method is conceptually simple, requires no training and can be implemented in a few lines of codes. On a number of image generation tasks we show that our results are often superior to single-image-GANs, require no training, and can generate high quality images in a few seconds. Our implementation is available at https://github.com/ariel415el/GPDM

Key facts. Cited by 26 papers in the map (data through 7 Oct 2026). Builds on 16 of its 38 references that are in the map. First posted on arXiv 22 Mar 2022.

Builds on

16 of 38 references.

References · 16

  1. 2017Progressive Growing of GANs for Improved Quality, Stability, and Variationcited by 2,967
  2. 2017Billion-scale similarity search with GPUscited by 2,477
  3. 2017High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANscited by 1,176
  4. 2017Universal Style Transfer via Feature Transformscited by 352
  5. 2019SinGAN: Learning a Generative Model from a Single Natural Imagecited by 348
  6. 2017VEEGAN: Reducing Mode Collapse in GANs using Implicit Variational Learningcited by 199
  7. 2019Style Transfer by Relaxed Optimal Transport and Self-Similaritycited by 167
  8. 2017Demystifying Neural Style Transfercited by 152
  9. 2018The Contextual Loss for Image Transformation with Non-Aligned Datacited by 135
  10. 2018Generative Modeling using the Sliced Wasserstein Distancecited by 108
  11. 2016Combining Markov Random Fields and Convolutional Neural Networks for Image Synthesiscited by 102
  12. 2020Improved Techniques for Training Single-Image GANscited by 77
4 more
  1. 2021Drop the GAN: In Defense of Patches Nearest Neighbors as Single Image Generative Modelscited by 51
  2. 2018Sliced-Wasserstein Autoencoder: An Embarrassingly Simple Generative Modelcited by 33
  3. 2019Wasserstein Style Transfercited by 21
  4. 2021Generating Novel Scene Compositions from Single Images and Videoscited by 8

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 3; 2023: 8; 2024: 2; 2025: 11; 2026: 2.

26 papers.

Sort:

By area:

  1. 2022SinDDM: A Single Image Denoising Diffusion Modelcited by 60
  2. 2025Attention Distillation: A Unified Approach to Visual Characteristics Transfercited by 32
  3. 2023Patch-based 3D Natural Scene Generation from a Single Examplecited by 15
  4. 2024Multiscale Sliced Wasserstein Distances as Perceptual Color Difference Measurescited by 14
  5. 2023Image Synthesis under Limited Data: A Survey and Taxonomycited by 9
  6. 2023Learning from small data sets: Patch-based regularizers in inverse problems for image reconstructioncited by 9
  7. 2023Wasserstein Distortion: Unifying Fidelity and Realismcited by 8
  8. 2023High-Fidelity Zero-Shot Texture Anomaly Localization Using Feature Correspondence Analysiscited by 7
  9. 2025A User's Guide to Sampling Strategies for Sliced Optimal Transportcited by 7
  10. 2024Misalignment-Robust Frequency Distribution Loss for Image Transformationcited by 6
  11. 2023Patch of Invisibility: Naturalistic Physical Black-Box Adversarial Attacks on Object Detectorscited by 5
  12. 2025Slicing Wasserstein Over Wasserstein Via Functional Optimal Transportcited by 5
14 more
  1. 2025ShapeShifter: 3D Variations Using Multiscale and Sparse Point-Voxel Diffusioncited by 3
  2. 2022OAIR: Object-Aware Image Retargeting Using PSO and Aesthetic Quality Assessmentcited by 1
  3. 2022Time Series Synthesis via Multi-scale Patch-based Generation of Wavelet Scalogramcited by 1
  4. 2025Video Consistency Distance: Enhancing Temporal Consistency for Image-to-Video Generation via Reward-Based Fine-Tuningcited by 1
  5. 2025Example-Based Feature Painting on Texturescited by 1
  6. 2026Efficient and Training-Free Single-Image Diffusion Modelscited by 1
  7. 2023One-shot Generative Distribution Matching for Augmented RF-based UAV Identificationcited by 0
  8. 2023Supervised Deep Learning for Content-Aware Image Retargeting with Fourier Convolutionscited by 0
  9. 2025HALO: Human-Aligned End-to-end Image Retargeting with Layered Transformationscited by 0
  10. 2025ReSWD: ReSTIR'd, not shaken. Combining Reservoir Sampling and Sliced Wasserstein Distance for Variance Reductioncited by 0
  11. 2025Quantized FCA: Efficient Zero-Shot Texture Anomaly Detectioncited by 0
  12. 2025Object-IR: Leveraging Object Consistency and Mesh Deformation for Self-Supervised Image Retargetingcited by 0
  13. 2025SinSEMI: A One-Shot Image Generation Model and Data-Efficient Evaluation Framework for Semiconductor Inspection Equipmentcited by 0
  14. 2026FRPSS: Feature Rearrangement in Pre-Shape Space for Single-Image Generationcited by 0