mlsanityData to 7 Oct 2026

FS-NCSR: Increasing Diversity of the Super-Resolution Space via Frequency Separation and Noise-Conditioned Normalizing Flow

arXiv 2204.09679 ↗First posted 20 Apr 2022cs.CV cs.LG eess.IVcited by 8

Not on the map. Why?

Abstract (arXiv)

Super-resolution suffers from an innate ill-posed problem that a single low-resolution (LR) image can be from multiple high-resolution (HR) images. Recent studies on the flow-based algorithm solve this ill-posedness by learning the super-resolution space and predicting diverse HR outputs. Unfortunately, the diversity of the super-resolution outputs is still unsatisfactory, and the outputs from the flow-based model usually suffer from undesired artifacts which causes low-quality outputs. In this paper, we propose FS-NCSR which produces diverse and high-quality super-resolution outputs using frequency separation and noise conditioning compared to the existing flow-based approaches. As the sharpness and high-quality detail of the image rely on its high-frequency information, FS-NCSR only estimates the high-frequency information of the high-resolution outputs without redundant low-frequency components. Through this, FS-NCSR significantly improves the diversity score without significant image quality degradation compared to the NCSR, the winner of the previous NTIRE 2021 challenge.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 26 of its 41 references that are in the map. First posted on arXiv 20 Apr 2022.

Builds on

26 of 41 references.

References · 26

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  3. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  4. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
  5. 2022Generative Adversarial Networkscited by 5,603
  6. 2019Generative Modeling by Estimating Gradients of the Data Distributioncited by 4,067
  7. 2015Variational Inference with Normalizing Flowscited by 2,161
  8. 2016Density estimation using Real NVPcited by 2,029
  9. 2016Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Networkcited by 1,898
  10. 2017Enhanced Deep Residual Networks for Single Image Super-Resolutioncited by 1,543
  11. 2018Glow: Generative Flow with Invertible 1x1 Convolutionscited by 1,539
  12. 2018Image Super-Resolution Using Very Deep Residual Channel Attention Networkscited by 1,256
14 more
  1. 2018ESRGAN: Enhanced Super-Resolution Generative Adversarial Networkscited by 1,250
  2. 2014Image Super-Resolution Using Deep Convolutional Networkscited by 1,218
  3. 2014NICE: Non-linear Independent Components Estimationcited by 1,052
  4. 2015Accurate Image Super-Resolution Using Very Deep Convolutional Networkscited by 902
  5. 2018Residual Dense Network for Image Super-Resolutioncited by 796
  6. 2021Resolution-robust Large Mask Inpainting with Fourier Convolutionscited by 730
  7. 2019Flow++: Improving Flow-Based Generative Models with Variational Dequantization and Architecture Designcited by 258
  8. 2021Deblurring via Stochastic Refinementcited by 248
  9. 2020SRFlow: Learning the Super-Resolution Space with Normalizing Flowcited by 209
  10. 2020SoftFlow: Probabilistic Framework for Normalizing Flow on Manifoldscited by 97
  11. 2021Hierarchical Conditional Flow: A Unified Framework for Image Super-Resolution and Image Rescalingcited by 73
  12. 2018Efficient and Accurate MRI Super-Resolution using a Generative Adversarial Network and 3D Multi-Level Densely Connected Networkcited by 64
  13. 2020ESRGAN+ : Further Improving Enhanced Super-Resolution Generative Adversarial Networkcited by 27
  14. 2021Noise Conditional Flow Model for Learning the Super-Resolution Spacecited by 19

Built on by

Cited by year

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

8 papers.

  1. 2023A Conditional Normalizing Flow for Accelerated Multi-Coil MR Imagingcited by 10
  2. 2024NAF-DPM: A Nonlinear Activation-Free Diffusion Probabilistic Model for Document Enhancementcited by 9
  3. 2022On the Robustness of Normalizing Flows for Inverse Problems in Imagingcited by 8
  4. 2024Boosting Flow-based Generative Super-Resolution Models via Learned Priorcited by 7
  5. 2023BlindHarmony: "Blind" Harmonization for MR Images via Flow modelcited by 5
  6. 2024Adaptive Selection of Sampling-Reconstruction in Fourier Compressed Sensingcited by 2
  7. 2025Leveraging Vision-Language Models to Select Trustworthy Super-Resolution Samples Generated by Diffusion Modelscited by 2
  8. 2024Trustworthy SR: Resolving Ambiguity in Image Super-resolution via Diffusion Models and Human Feedbackcited by 1