mlsanityData to 14 Oct 2026

Physics-Informed Neural Network Super Resolution for Advection-Diffusion Models

arXiv 2011.02519 ↗First posted 4 Nov 2020cs.CV physics.geo-phcited by 16

On the map: Diffusion models → Flow matching, time series and science

Abstract (arXiv)

Physics-informed neural networks (NN) are an emerging technique to improve spatial resolution and enforce physical consistency of data from physics models or satellite observations. A super-resolution (SR) technique is explored to reconstruct high-resolution images (4×) from lower resolution images in an advection-diffusion model of atmospheric pollution plumes. SR performance is generally increased when the advection-diffusion equation constrains the NN in addition to conventional pixel-based constraints. The ability of SR techniques to also reconstruct missing data is investigated by randomly removing image pixels from the simulations and allowing the system to learn the content of missing data. Improvements in S/N of 11% are demonstrated when physics equations are included in SR with 40% pixel loss. Physics-informed NNs accurately reconstruct corrupted images and generate better results compared to the standard SR approaches.

Key facts. Cited by 16 papers in the map (data through 14 Oct 2026). Placed in Flow matching, time series and science, one of 14 directions in Diffusion models. Builds on 8 of its 22 references that are in the map. First posted on arXiv 4 Nov 2020.

Builds on

8 of 22 references.

References · 8

  1. 2016Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Networkcited by 1,908
  2. 2017Enhanced Deep Residual Networks for Single Image Super-Resolutioncited by 1,552
  3. 2018ESRGAN: Enhanced Super-Resolution Generative Adversarial Networkscited by 1,254
  4. 2015Accurate Image Super-Resolution Using Very Deep Convolutional Networkscited by 903
  5. 2015Deeply-Recursive Convolutional Network for Image Super-Resolutioncited by 317
  6. 2017MemNet: A Persistent Memory Network for Image Restorationcited by 244
  7. 2020MeshfreeFlowNet: A Physics-Constrained Deep Continuous Space-Time Super-Resolution Frameworkcited by 78
  8. 2019Using Physics-Informed Super-Resolution Generative Adversarial Networks for Subgrid Modeling in Turbulent Reactive Flowscited by 8

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 2; 2022: 4; 2023: 4; 2024: 1; 2025: 4; 2026: 1.

16 papers.

Sort:

By area:

  1. 2023Physics-Informed Computer Vision: A Review and Perspectivescited by 28
  2. 2023SuperBench: A Super-Resolution Benchmark Dataset for Scientific Machine Learningcited by 18
  3. 2021Adversarial sampling of unknown and high-dimensional conditional distributionscited by 13
  4. 2021S3RP: Self-Supervised Super-Resolution and Prediction for Advection-Diffusion Processcited by 9
  5. 2022Physics-informed Deep Super-resolution for Spatiotemporal Datacited by 9
  6. 2025Representing Flow Fields with Divergence-Free Kernels for Reconstructioncited by 5
  7. 2022Rotationally Equivariant Super-Resolution of Velocity Fields in Two-Dimensional Fluids Using Convolutional Neural Networkscited by 3
  8. 2023Filtered Partial Differential Equations: a robust surrogate constraint in physics-informed deep learning frameworkcited by 1
  9. 2025Wasserstein GAN-Based Precipitation Downscaling with Optimal Transport for Enhancing Perceptual Realismcited by 1
  10. 2025Probabilistic Super-Resolution for Urban Micrometeorology via a Schr"odinger Bridgecited by 1
  11. 2026Physics-Informed Video Diffusion For Shallow Water Equationscited by 1
  12. 2022Tools for Extracting Spatio-Temporal Patterns in Meteorological Image Sequences: From Feature Engineering to Attention-Based Neural Networkscited by 0
4 more
  1. 2022Multi-task Learning for Source Attribution and Field Reconstruction for Methane Monitoringcited by 0
  2. 2023An Operator Learning Framework for Spatiotemporal Super-resolution of Scientific Simulationscited by 0
  3. 2024How to Re-enable PDE Loss for Physical Systems Modeling Under Partial Observationcited by 0
  4. 2025Atmos-Bench: 3D Atmospheric Structures for Climate Insightcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Enhancing Weather Predictions: Super-Resolution via Deep Diffusion Models
  2. 2025Multimodal Atmospheric Super-Resolution With Deep Generative Models
  3. 2024PiRD: Physics-informed Residual Diffusion for Flow Field Reconstruction
  4. 2024Stochastic Flow Matching for Resolving Small-Scale Physics
  5. 2025A PDE-Informed Latent Diffusion Model for 2-m Temperature Downscaling
  6. 2026Physics-Consistent Diffusion for Efficient Fluid Super-Resolution via Multiscale Residual Correction
  7. 2023EDiffSR: An Efficient Diffusion Probabilistic Model for Remote Sensing Image Super-Resolution
  8. 2025NeurOp-Diff:Continuous Remote Sensing Image Super-Resolution via Neural Operator Diffusion