mlsanityData to 14 Oct 2026

Visual Anagrams: Generating Multi-View Optical Illusions with Diffusion Models

arXiv 2311.17919 ↗First posted 29 Nov 2023cs.CVcited by 46

On the map: Diffusion models → Image generation, text-to-image and editing

Abstract (arXiv)

We address the problem of synthesizing multi-view optical illusions: images that change appearance upon a transformation, such as a flip or rotation. We propose a simple, zero-shot method for obtaining these illusions from off-the-shelf text-to-image diffusion models. During the reverse diffusion process, we estimate the noise from different views of a noisy image, and then combine these noise estimates together and denoise the image. A theoretical analysis suggests that this method works precisely for views that can be written as orthogonal transformations, of which permutations are a subset. This leads to the idea of a visual anagram--an image that changes appearance under some rearrangement of pixels. This includes rotations and flips, but also more exotic pixel permutations such as a jigsaw rearrangement. Our approach also naturally extends to illusions with more than two views. We provide both qualitative and quantitative results demonstrating the effectiveness and flexibility of our method. Please see our project webpage for additional visualizations and results: https://dangeng.github.io/visual_anagrams/

Key facts. Cited by 46 papers in the map (data through 14 Oct 2026). Placed in Image generation, text-to-image and editing, one of 14 directions in Diffusion models. Builds on 24 of its 48 references that are in the map. First posted on arXiv 29 Nov 2023.

Builds on

24 of 48 references.

Closest ancestors · 3

Same direction, newest first

  1. 2023Adding Conditional Control to Text-to-Image Diffusion Models10 Feb
  2. 2022Compositional Visual Generation with Composable Diffusion Models3 Jun
  3. 2021GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models20 Dec

Other references · 21

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  4. 2020Denoising Diffusion Implicit Modelscited by 9,781
  5. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 9,122
  6. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,255
  7. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  8. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,058
  9. 2022Classifier-Free Diffusion Guidancecited by 5,847
  10. 2021Zero-Shot Text-to-Image Generationcited by 4,288
  11. 2022DreamFusion: Text-to-3D using 2D Diffusioncited by 3,025
  12. 2022Score Jacobian Chaining: Lifting Pretrained 2D Diffusion Models for 3D Generationcited by 597
9 more
  1. 2022Diffusion models as plug-and-play priorscited by 231
  2. 2023Reduce, Reuse, Recycle: Compositional Generation with Energy-Based Diffusion Models and MCMCcited by 220
  3. 2018Adversarial Examples that Fool both Computer Vision and Time-Limited Humanscited by 73
  4. 2021Learning to Compose Visual Relationscited by 72
  5. 2023Intriguing properties of generative classifierscited by 52
  6. 2023Compositional Sculpting of Iterative Generative Processescited by 27
  7. 2022Designing Perceptual Puzzles by Differentiating Probabilistic Programscited by 12
  8. 2020Toward Quantifying Ambiguities in Artistic Imagescited by 8
  9. 2022GANmouflage: 3D Object Nondetection with Texture Fieldscited by 7

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 1; 2024: 14; 2025: 16; 2026: 15.

46 papers.

Sort:

By area:

  1. 2024Video-Guided Foley Sound Generation with Multimodal Controlscited by 49
  2. 2024StyleMaster: Stylize Your Video with Artistic Generation and Translationcited by 47
  3. 2024Rethinking Score Distillation as a Bridge Between Image Distributionscited by 38
  4. 2023Diffusion Illusions: Hiding Images in Plain Sightcited by 26
  5. 2024SyncTweedies: A General Generative Framework Based on Synchronized Diffusionscited by 26
  6. 2025Lossy Compression with Pretrained Diffusion Modelscited by 25
  7. 2024Factorized Diffusion: Perceptual Illusions by Noise Decompositioncited by 23
  8. 2024Images that Sound: Composing Images and Sounds on a Single Canvascited by 17
  9. 2024Diffusion-Based Visual Art Creation: A Survey and New Perspectivescited by 17
  10. 2024HiPrompt: Tuning-free Higher-Resolution Generation with Hierarchical MLLM Promptscited by 15
  11. 2024Illusory VQA: Benchmarking and Enhancing Multimodal Models on Visual Illusionscited by 11
  12. 2025Anchor Token Matching: Implicit Structure Locking for Training-free AR Image Editingcited by 10
34 more
  1. 2024The Art of Deception: Color Visual Illusions and Diffusion Modelscited by 9
  2. 2025StochSync: Stochastic Diffusion Synchronization for Image Generation in Arbitrary Spacescited by 8
  3. 2025Unconditional Priors Matter! Improving Conditional Generation of Fine-Tuned Diffusion Modelscited by 8
  4. 2025PTDiffusion: Free Lunch for Generating Optical Illusion Hidden Pictures with Phase-Transferred Diffusion Modelcited by 7
  5. 2025Leveraging Semantic Attribute Binding for Free-Lunch Color Control in Diffusion Modelscited by 7
  6. 2025LookingGlass: Generative Anamorphoses via Laplacian Pyramid Warpingcited by 7
  7. 2024Diffusion-based Visual Anagram as Multi-task Learningcited by 6
  8. 2024MatchDiffusion: Training-free Generation of Match-cutscited by 5
  9. 2024Illusion3D: 3D Multiview Illusion with 2D Diffusion Priorscited by 5
  10. 2025GPS as a Control Signal for Image Generationcited by 5
  11. 2025MOSAIC: Generating Consistent, Privacy-Preserving Scenes from Multiple Depth Views in Multi-Room Environmentscited by 5
  12. 2025Hand-Shadow Posercited by 4
  13. 2025SyncSDE: A Probabilistic Framework for Diffusion Synchronizationcited by 3
  14. 2025Visual Anagrams Reveal Hidden Differences in Holistic Shape Processing Across Vision Modelscited by 3
  15. 2025PickStyle: Video-to-Video Style Transfer with Context-Style Adapterscited by 3
  16. 2026Stroke of Surprise: Progressive Semantic Illusions in Vector Sketchingcited by 3
  17. 2025FreSca: Scaling in Frequency Space Enhances Diffusion Modelscited by 2
  18. 2025Refracting Reality: Generating Images with Realistic Transparent Objectscited by 2
  19. 2026ChArtist: Generating Pictorial Charts with Unified Spatial and Subject Controlcited by 2
  20. 2026SMSP: A Plug-and-Play Strategy of Multi-Scale Perception for MLLMs to Perceive Visual Illusionscited by 2
  21. 2026A Framework for Generating Semantically Ambiguous Images to Probe Human and Machine Perceptioncited by 1
  22. 2026Meschers: Geometry Processing of Impossible Objectscited by 1
  23. 2026SymTRELLIS: Symmetry-Enforced Voxel Latents for 3D Generationcited by 1
  24. 2026Asynchronous Multimodal Diffusion Policy Composition via Latency-Aware Guidance Fusioncited by 1
  25. 2026Now You See the Hate: Adaptive View Retrieval for Hidden Hateful Illusionscited by 1
  26. 2024Making Images from Images: Interleaving Denoising and Transformationcited by 0
  27. 2025StableMotion: One-Step Motion Estimation with Diffusion Priorcited by 0
  28. 2026Divide-and-Denoise: A Game-Theoretic Method for Fairly Composing Diffusion Modelscited by 0
  29. 2026Variational Test-time Optimization for Diffusion Synchronizationcited by 0
  30. 2026JanusMesh: Fast and Zero-Shot 3D Visual Illusion Generation via Cross-Space Denoisingcited by 0
  31. 2026PhotoQuilt: Training-Free Arbitrary-Resolution Photomosaics via Bootstrapped Tiled Denoisingcited by 0
  32. 2026Mirror Illusion Artcited by 0
  33. 2026Form and Void: Entangled Composition through an Autonomous AI Agentcited by 0
  34. 2026Moore, Escher, Penrose: A Conformal Golden Braidcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Structure-Semantic Co-optimized Latent Diffusion Model for Fast Visual Anagram Synthesis
  2. 2024Diffusion Models in 3D Vision: A Survey
  3. 2023Diffusion Models Generate Images Like Painters: an Analytical Theory of Outline First, Details Later
  4. 2023State of the Art on Diffusion Models for Visual Computing
  5. 2024Diff-2-in-1: Bridging Generation and Dense Perception with Diffusion Models
  6. 2024Diffusion Models in Low-Level Vision: A Survey
  7. 2023ViVid-1-to-3: Novel View Synthesis with Video Diffusion Models
  8. 2023ImageDream: Image-Prompt Multi-view Diffusion for 3D Generation