mlsanityData to 14 Oct 2026

FouriScale: A Frequency Perspective on Training-Free High-Resolution Image Synthesis

arXiv 2403.12963 ↗First posted 19 Mar 2024cs.CVcited by 59

On the map: Diffusion models → Diffusion foundations and sampling

Abstract (arXiv)

In this study, we delve into the generation of high-resolution images from pre-trained diffusion models, addressing persistent challenges, such as repetitive patterns and structural distortions, that emerge when models are applied beyond their trained resolutions. To address this issue, we introduce an innovative, training-free approach FouriScale from the perspective of frequency domain analysis. We replace the original convolutional layers in pre-trained diffusion models by incorporating a dilation technique along with a low-pass operation, intending to achieve structural consistency and scale consistency across resolutions, respectively. Further enhanced by a padding-then-crop strategy, our method can flexibly handle text-to-image generation of various aspect ratios. By using the FouriScale as guidance, our method successfully balances the structural integrity and fidelity of generated images, achieving an astonishing capacity of arbitrary-size, high-resolution, and high-quality generation. With its simplicity and compatibility, our method can provide valuable insights for future explorations into the synthesis of ultra-high-resolution images. The code will be released at https://github.com/LeonHLJ/FouriScale.

Key facts. Cited by 59 papers in the map (data through 14 Oct 2026). Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models. Builds on 42 of its 48 references that are in the map. First posted on arXiv 19 Mar 2024.

Builds on

42 of 48 references.

Closest ancestors · 14

Same direction, newest first

  1. 2024FiT: Flexible Vision Transformer for Diffusion Model19 Feb
  2. 2023ScaleCrafter: Tuning-free Higher-Resolution Visual Generation with Diffusion Models11 Oct
  3. 2023FreeU: Free Lunch in Diffusion U-Net20 Sep
  4. 2023Relay Diffusion: Unifying diffusion process across resolutions for image synthesis4 Sep
  5. 2023SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis4 Jul
  6. 2023Simple diffusion: End-to-end diffusion for high resolution images26 Jan
  7. 2023On the Importance of Noise Scheduling for Diffusion Models26 Jan
  8. 2022Scalable Diffusion Models with Transformers19 Dec
  9. 2022Classifier-Free Diffusion Guidance26 Jul
  10. 2021High-Resolution Image Synthesis with Latent Diffusion Models20 Dec
  11. 2021Cascaded Diffusion Models for High Fidelity Image Generation30 May
  12. 2021Diffusion Models Beat GANs on Image Synthesis11 May
2 more
  1. 2020Denoising Diffusion Implicit Models6 Oct
  2. 2020Denoising Diffusion Probabilistic Models19 Jun

Other references · 28

  1. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  2. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,355
  3. 2021Zero-Shot Text-to-Image Generationcited by 4,288
  4. 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,152
  5. 2022Prompt-to-Prompt Image Editing with Cross Attention Controlcited by 2,431
  6. 2023Align your Latents: High-Resolution Video Synthesis with Latent Diffusion Modelscited by 1,476
  7. 2018Demystifying MMD GANscited by 1,293
  8. 2018Image Super-Resolution Using Very Deep Residual Channel Attention Networkscited by 1,264
  9. 2022eDiff-I: Text-to-Image Diffusion Models with an Ensemble of Expert Denoiserscited by 931
  10. 2023MasaCtrl: Tuning-Free Mutual Self-Attention Control for Consistent Image Synthesis and Editingcited by 727
  11. 2021CogView: Mastering Text-to-Image Generation via Transformerscited by 694
  12. 2023AudioLDM: Text-to-Audio Generation with Latent Diffusion Modelscited by 599
16 more
  1. 2022LION: Latent Point Diffusion Models for 3D Shape Generationcited by 553
  2. 2023MultiDiffusion: Fusing Diffusion Paths for Controlled Image Generationcited by 536
  3. 2019Making Convolutional Networks Shift-Invariant Againcited by 391
  4. 2023Diffusion Self-Guidance for Controllable Image Generationcited by 305
  5. 2023Unleashing Text-to-Image Diffusion Models for Visual Perceptioncited by 269
  6. 2019Adaptive Attention Span in Transformerscited by 149
  7. 2023SyncDiffusion: Coherent Montage via Synchronized Joint Diffusionscited by 110
  8. 2023DemoFusion: Democratising High-Resolution Image Generation With No cited by 107
  9. 2015Spectral Representations for Convolutional Neural Networkscited by 81
  10. 2023Diffusion Model is Secretly a Training-free Open Vocabulary Semantic Segmentercited by 76
  11. 2023Mixture of Diffusers for scene composition and high resolution image generationcited by 56
  12. 2023Training-free Diffusion Model Adaptation for Variable-Sized Text-to-Image Synthesiscited by 56
  13. 2023ElasticDiffusion: Training-free Arbitrary Size Image Generation through Global-Local Content Separationcited by 37
  14. 2023Any-Size-Diffusion: Toward Efficient Text-Driven Synthesis for Any-Size HD Imagescited by 36
  15. 2022Learning strides in convolutional neural networkscited by 26
  16. 2023From Text to Mask: Localizing Entities Using the Attention of Text-to-Image Diffusion Modelscited by 16

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 17; 2025: 29; 2026: 13.

59 papers.

Sort:

By area:

  1. 2024Lumina-T2X: Transforming Text into Any Modality, Resolution, and Duration via Flow-based Large Diffusion Transformerscited by 137
  2. 2024Lumina-Next: Making Lumina-T2X Stronger and Faster with Next-DiTcited by 132
  3. 2024UltraPixel: Advancing Ultra-High-Resolution Image Synthesis to New Peakscited by 65
  4. 2024LinFusion: 1 GPU, 1 Minute, 16K Imagecited by 43
  5. 2024DiffuseHigh: Training-free Progressive High-Resolution Image Synthesis through Structure Guidancecited by 34
  6. 2024FreeScale: Unleashing the Resolution of Diffusion Models via Tuning-Free Scale Fusioncited by 30
  7. 2024CLEAR: Conv-Like Linearization Revs Pre-Trained Diffusion Transformers Upcited by 28
  8. 2024MegaFusion: Extend Diffusion Models towards Higher-resolution Image Generation without Further Tuningcited by 25
  9. 2025Latent Space Super-Resolution for Higher-Resolution Image Generation with Diffusion Modelscited by 25
  10. 2024I-Max: Maximize the Resolution Potential of Pre-trained Rectified Flow Transformers with Projected Flowcited by 23
  11. 2025HiFlow: Training-free High-Resolution Image Generation with Flow-Aligned Guidancecited by 22
  12. 2025UltraHR-100K: Enhancing UHR Image Synthesis with A Large-Scale High-Quality Datasetcited by 18
38 more
  1. 2024FreCaS: Efficient Higher-Resolution Image Generation via Frequency-aware Cascaded Samplingcited by 16
  2. 2024FRAG: Frequency Adapting Group for Diffusion Video Editingcited by 15
  3. 2024HiPrompt: Tuning-free Higher-Resolution Generation with Hierarchical MLLM Promptscited by 15
  4. 2025MagicTryOn: Harnessing Diffusion Transformer for Garment-Preserving Video Virtual Try-oncited by 15
  5. 2024Is One GPU Enough? Pushing Image Generation at Higher-Resolutions with Foundation Modelscited by 14
  6. 2025DyPE: Dynamic Position Extrapolation for Ultra High Resolution Diffusioncited by 14
  7. 2025Ultra-Resolution Adaptation with Easecited by 13
  8. 2024FAM Diffusion: Frequency and Attention Modulation for High-Resolution Image Generation with Stable Diffusioncited by 12
  9. 2025UltraImage: Rethinking Resolution Extrapolation in Image Diffusion Transformerscited by 12
  10. 2025HiStream: Efficient High-Resolution Video Generation via Redundancy-Eliminated Streamingcited by 8
  11. 2025Latent Wavelet Diffusion For Ultra-High-Resolution Image Synthesiscited by 7
  12. 2025Decoupling High and Low Frequencies for Faithful Image Generation with Fine Detailscited by 7
  13. 2024RepLDM: Reprogramming Pretrained Latent Diffusion Models for High-Quality, High-Efficiency, High-Resolution Image Generationcited by 6
  14. 2025Revisiting Diffusion Models: From Generative Pre-training to One-Step Generationcited by 6
  15. 2025ScaleDiff: Higher-Resolution Image Synthesis via Efficient and Model-Agnostic Diffusioncited by 6
  16. 2025UltraFlux: Data-Model Co-Design for High-quality Native 4K Text-to-Image Generation across Diverse Aspect Ratioscited by 6
  17. 2026Frequency-Aware Flow Matching for High-Quality Image Generationcited by 6
  18. 2025HiWave: Training-Free High-Resolution Image Generation via Wavelet-Based Diffusion Samplingcited by 5
  19. 2025Parallel Sequence Modeling via Generalized Spatial Propagation Networkcited by 4
  20. 2025InfGen: A Resolution-Agnostic Paradigm for Scalable Image Synthesiscited by 4
  21. 2024GaussianPainter: Painting Point Cloud into 3D Gaussians with Normal Guidancecited by 3
  22. 2025LEDiT: Your Length-Extrapolatable Diffusion Transformer without Positional Encodingcited by 3
  23. 2025Exploring Position Encoding in Diffusion U-Net for Training-free High-resolution Image Generationcited by 3
  24. 2025UltraImageGen: Efficient Ultra-High-Resolution Image Generation with Hierarchical Local Attentioncited by 3
  25. 2026SEGA: Spectral-Energy Guided Attention for Resolution Extrapolation in Diffusion Transformerscited by 3
  26. 2025Creatively Upscaling Images with Global-Regional Priorscited by 2
  27. 2025Low-Resolution Editing is All You Need for High-Resolution Editingcited by 2
  28. 2025FRAMER: Frequency-Aligned Self-Distillation with Adaptive Modulation Leveraging Diffusion Priors for Real-World Image Super-Resolutioncited by 2
  29. 2025ResDiT: Evoking the Intrinsic Resolution Scalability in Diffusion Transformerscited by 2
  30. 2024Improvements to SDXL in NovelAI Diffusion V3cited by 1
  31. 2025RectifiedHR: Enable Efficient High-Resolution Synthesis via Energy Rectificationcited by 1
  32. 2025CineScale: Tuning-Free High-Resolution Video Generationcited by 1
  33. 2025InfoScale: Unleashing Training-free Variable-scaled Image Generation via Effective Utilization of Informationcited by 1
  34. 2026PixelRush: Ultra-Fast, Training-Free High-Resolution Image Generation via One-step Diffusioncited by 1
  35. 2026PixVerve: Advancing Native UHR Image Generation to 100MP with a Large-Scale High-Quality Datasetcited by 1
  36. 2026Balancing Frequencies and Pixels in Flow Matchingcited by 1
  37. 2025APT: Improving Diffusion Models for High Resolution Image Generation with Adaptive Path Tracingcited by 0
  38. 2025NoiseShift: Resolution-Aware Noise Recalibration for Better Low-Resolution Image Generationcited by 0

50 most cited of 59 All 59 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025Ultra-High-Resolution Image Synthesis: Data, Method and Evaluation
  2. 2025Diffusion-4K: Ultra-High-Resolution Image Synthesis with Latent Diffusion Models
  3. 2025Native-Resolution Image Synthesis
  4. 2023Dissecting Arbitrary-scale Super-resolution Capability from Pre-trained Diffusion Generative Models
  5. 2024ResMaster: Mastering High-Resolution Image Generation via Structural and Fine-Grained Guidance
  6. 2025Local-Global Context-Aware and Structure-Preserving Image Super-Resolution
  7. 2026Cross-Resolution Diffusion Models via Network Pruning
  8. 2024Make a Cheap Scaling: A Self-Cascade Diffusion Model for Higher-Resolution Adaptation