mlsanityData to 7 Oct 2026

EpiDiff: Enhancing Multi-View Synthesis via Localized Epipolar-Constrained Diffusion

arXiv 2312.06725 ↗First posted 11 Dec 2023cs.CVcited by 72

On the map: Diffusion models → 3D, point cloud and scene generation

Abstract (arXiv)

Generating multiview images from a single view facilitates the rapid generation of a 3D mesh conditioned on a single image. Recent methods that introduce 3D global representation into diffusion models have shown the potential to generate consistent multiviews, but they have reduced generation speed and face challenges in maintaining generalizability and quality. To address this issue, we propose EpiDiff, a localized interactive multiview diffusion model. At the core of the proposed approach is to insert a lightweight epipolar attention block into the frozen diffusion model, leveraging epipolar constraints to enable cross-view interaction among feature maps of neighboring views. The newly initialized 3D modeling module preserves the original feature distribution of the diffusion model, exhibiting compatibility with a variety of base diffusion models. Experiments show that EpiDiff generates 16 multiview images in just 12 seconds, and it surpasses previous methods in quality evaluation metrics, including PSNR, SSIM and LPIPS. Additionally, EpiDiff can generate a more diverse distribution of views, improving the reconstruction quality from generated multiviews. Please see our project page at https://huanngzh.github.io/EpiDiff/.

Key facts. Cited by 72 papers in the map (data through 7 Oct 2026). Placed in 3D, point cloud and scene generation, one of 14 directions in Diffusion models. Builds on 69 of its 79 references that are in the map. First posted on arXiv 11 Dec 2023.

Builds on

69 of 79 references.

Closest ancestors · 34

Same direction, newest first

  1. 2023Text-to-3D with Classifier Score Distillation30 Oct
  2. 2023DreamCraft3D: Hierarchical 3D Generation with Bootstrapped Diffusion Prior25 Oct
  3. 2023HiFi-123: Towards High-fidelity One Image to 3D Content Generation10 Oct
  4. 2023Large-Vocabulary 3D Diffusion Model with Transformer14 Sep
  5. 2023Points-to-3D: Bridging the Gap between Sparse Points and Shape-Controllable Text-to-3D Generation26 Jul
  6. 2023AutoDecoding Latent 3D Diffusion Models7 Jul
  7. 2023Magic123: One Image to High-Quality 3D Object Generation Using Both 2D and 3D Diffusion Priors30 Jun
  8. 2023One-2-3-45: Any Single Image to 3D Mesh in 45 Seconds without Per-Shape Optimization29 Jun
  9. 2023ProlificDreamer: High-Fidelity and Diverse Text-to-3D Generation with Variational Score Distillation25 May
  10. 2023TextMesh: Generation of Realistic 3D Meshes From Text Prompts24 Apr
  11. 2023Long-Term Photometric Consistent Novel View Synthesis with Diffusion Models21 Apr
  12. 2023NeuralField-LDM: Scene Generation with Hierarchical Latent Diffusion Models19 Apr
22 more
  1. 2023Single-Stage Diffusion NeRF: A Unified Approach to 3D Generation and Reconstruction13 Apr
  2. 2023Control3Diff: Learning Controllable 3D Diffusion Models from Single-view Images13 Apr
  3. 2023Generative Novel View Synthesis with 3D-Aware Diffusion Models5 Apr
  4. 2023DreamAvatar: Text-and-Shape Guided 3D Human Avatar Generation via Diffusion Models3 Apr
  5. 20233D-aware Image Generation using 2D Diffusion Models31 Mar
  6. 2023Consistent View Synthesis with Pose-Guided Diffusion Models30 Mar
  7. 2023Make-It-3D: High-Fidelity 3D Creation from A Single Image with Diffusion Prior24 Mar
  8. 2023Fantasia3D: Disentangling Geometry and Appearance for High-quality Text-to-3D Content Creation24 Mar
  9. 2023Zero-1-to-3: Zero-shot One Image to 3D Object20 Mar
  10. 2023Let 2D Diffusion Model Know 3D-Consistency for Robust Text-to-3D Generation14 Mar
  11. 20233DGen: Triplane Latent Diffusion for Textured Mesh Generation9 Mar
  12. 2022Point-E: A System for Generating 3D Point Clouds from Complex Prompts16 Dec
  13. 2022Rodin: A Generative Model for Sculpting 3D Digital Avatars Using Diffusion12 Dec
  14. 2022NeRDi: Single-View NeRF Synthesis with Language-Guided Diffusion as General Image Priors6 Dec
  15. 2022DiffRF: Rendering-Guided 3D Radiance Field Diffusion2 Dec
  16. 2022Score Jacobian Chaining: Lifting Pretrained 2D Diffusion Models for 3D Generation1 Dec
  17. 2022Magic3D: High-Resolution Text-to-3D Content Creation18 Nov
  18. 2022RenderDiffusion: Image Diffusion for 3D Reconstruction, Inpainting and Generation17 Nov
  19. 2022LION: Latent Point Diffusion Models for 3D Shape Generation12 Oct
  20. 2022Novel View Synthesis with Diffusion Models6 Oct
  21. 2022DreamFusion: Text-to-3D using 2D Diffusion29 Sep
  22. 2021Diffusion Probabilistic Models for 3D Point Cloud Generation2 Mar

Other references · 35

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  3. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  4. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,539
  5. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,322
  6. 2021NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstructioncited by 1,701
  7. 2020pixelNeRF: Neural Radiance Fields from One or Few Imagescited by 1,574
  8. 2022Objaverse: A Universe of Annotated 3D Objectscited by 1,522
  9. 2021MVSNeRF: Fast Generalizable Radiance Field Reconstruction from Multi-View Stereocited by 735
  10. 2021IBRNet: Learning Multi-View Image-Based Renderingcited by 698
  11. 2021Common Objects in 3D: Large-Scale Learning and Evaluation of Real-life 3D Category Reconstructioncited by 646
  12. 2022Google Scanned Objects: A High-Quality Dataset of 3D Scanned Household Itemscited by 601
23 more
  1. 2021Zero-Shot Text-Guided Object Generation with Dream Fieldscited by 513
  2. 2022GET3D: A Generative Model of High Quality 3D Textured Shapes Learned from Imagescited by 446
  3. 2021Light Field Networks: Neural Scene Representations with Single-Evaluation Renderingcited by 330
  4. 2023RealFusion: 360deg Reconstruction of Any Object from a Single Imagecited by 305
  5. 2022SDFusion: Multimodal 3D Shape Completion, Reconstruction, and Generationcited by 287
  6. 2023DreamBooth3D: Subject-Driven Text-to-3D Generationcited by 235
  7. 2021Scene Representation Transformer: Geometry-Free Novel View Synthesis Through Set-Latent Scene Representationscited by 222
  8. 2023MVDiffusion: Enabling Holistic Multi-view Image Generation with Correspondence-Aware Diffusioncited by 210
  9. 2023MeshDiffusion: Score-based Generative 3D Mesh Modelingcited by 198
  10. 2021GeoNeRF: Generalizing NeRF with Geometry Priorscited by 195
  11. 2021Neural Rays for Occlusion-aware Image-based Renderingcited by 192
  12. 2023NerfDiff: Single-image View Synthesis with NeRF-guided Distillation from 3D-aware Diffusioncited by 179
  13. 2022SparseNeuS: Fast Generalizable Neural Surface Reconstruction from Sparse Viewscited by 175
  14. 2022Generalizable Patch-Based Neural Renderingcited by 121
  15. 2022NeRFusion: Fusing Radiance Fields for Large-Scale Scene Reconstructioncited by 99
  16. 2023Anything-3D: Towards Single-view Anything Reconstruction in the Wildcited by 91
  17. 2022Object Scene Representation Transformercited by 90
  18. 2022ViewFormer: NeRF-free Neural Rendering from Few Images Using Transformerscited by 70
  19. 2021Unsupervised Learning of 3D Object Categories from Videos in the Wildcited by 67
  20. 2023DITTO-NeRF: Diffusion-based Iterative Text To Omni-directional 3D Modelcited by 43
  21. 2023ContraNeRF: Generalizable Neural Radiance Fields for Synthetic-to-real Novel View Synthesis via Contrastive Learningcited by 14
  22. 2023Pushing the Limits of 3D Shape Generation at Scalecited by 11
  23. 2023SparseGNV: Generating Novel Views of Indoor Scenes with Sparse Input Viewscited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 1; 2024: 23; 2025: 33; 2026: 15.

72 papers.

Sort:

By area:

  1. 2023Triplane Meets Gaussian Splatting: Fast and Generalizable Single-View 3D Reconstruction with Transformerscited by 257
  2. 2024MIDI: Multi-Instance Diffusion for Single Image to 3D Scene Generationcited by 113
  3. 2024MV-Adapter: Multi-view Consistent Image Generation Made Easycited by 111
  4. 2025MEt3R: Measuring Multi-View Consistency in Generated Imagescited by 91
  5. 2025SparseFlex: High-Resolution and Arbitrary-Topology 3D Shape Modelingcited by 76
  6. 2025OmniPart: Part-Aware 3D Generation with Semantic Decoupling and Structural Cohesioncited by 60
  7. 2024GVGEN: Text-to-3D Generation with Volumetric Representationcited by 51
  8. 2024Human-3Diffusion: Realistic Avatar Creation via Explicit 3D Consistent Diffusion Modelscited by 44
  9. 2025VoxHammer: Training-Free Precise and Coherent 3D Editing in Native 3D Spacecited by 40
  10. 2024DiffPano: Scalable and Consistent Text to Panorama Generation with Spherical Epipolar-Aware Diffusioncited by 38
  11. 2024MVD-Fusion: Single-view 3D via Depth-consistent Multi-view Generationcited by 35
  12. 2025CubeDiff: Repurposing Diffusion-Based Image Models for Panorama Generationcited by 35
38 more
  1. 2024Ouroboros3D: Image-to-3D Generation via 3D-aware Recursive Diffusioncited by 31
  2. 2025MaterialMVP: Illumination-Invariant Material Generation via Multi-view PBR Diffusioncited by 28
  3. 2024Hi3D: Pursuing High-Resolution Image-to-3D Generation with Video Diffusion Modelscited by 27
  4. 2024Cavia: Camera-controllable Multi-view Video Diffusion with View-Integrated Attentioncited by 24
  5. 2025DepR: Depth Guided Single-view Scene Reconstruction with Instance-level Diffusioncited by 23
  6. 2025Pippo: High-Resolution Multi-View Humans from a Single Imagecited by 20
  7. 20243D-Adapter: Geometry-Consistent Multi-View Diffusion for High-Quality 3D Generationcited by 19
  8. 2025GAS: Generative Avatar Synthesis from a Single Imagecited by 19
  9. 2024From Parts to Whole: A Unified Reference Framework for Controllable Human Image Generationcited by 17
  10. 2024PCDreamer: Point Cloud Completion Through Multi-view Diffusion Priorscited by 16
  11. 2024TAR3D: Creating High-Quality 3D Assets via Next-Part Predictioncited by 16
  12. 2024GenesisTex: Adapting Image Denoising Diffusion to Texture Spacecited by 15
  13. 2025Survey of Video Diffusion Models: Foundations, Implementations, and Applicationscited by 14
  14. 2025VideoPanda: Video Panoramic Diffusion with Multi-view Attentioncited by 13
  15. 20243DEnhancer: Consistent Multi-View Diffusion for 3D Enhancementcited by 12
  16. 2025Fast Multi-view Consistent 3D Editing with Video Priorscited by 12
  17. 2024HouseCrafter: Lifting Floorplans to 3D Scenes with 2D Diffusion Modelcited by 9
  18. 2024MOVIS: Enhancing Multi-Object Novel View Synthesis for Indoor Scenescited by 9
  19. 2025MVGBench: Comprehensive Benchmark for Multi-view Generation Modelscited by 9
  20. 2026SegviGen: Repurposing 3D Generative Model for Part Segmentationcited by 9
  21. 2025MEAT: Multiview Diffusion Model for Human Generation on Megapixels with Mesh Attentioncited by 8
  22. 2026PhysForge: Generating Physics-Grounded 3D Assets for Interactive Virtual Worldcited by 8
  23. 2024DreamCraft3D++: Efficient Hierarchical 3D Generation with Multi-Plane Reconstruction Modelcited by 7
  24. 2025ExploreGS: Explorable 3D Scene Reconstruction with Virtual Camera Samplings and Diffusion Priorscited by 6
  25. 2024GeoGS3D: Single-view 3D Reconstruction via Geometric-aware Diffusion Model and Gaussian Splattingcited by 5
  26. 2024StdGEN: Semantic-Decomposed 3D Character Generation from Single Imagescited by 5
  27. 2025CoreEditor: Correspondence-constrained Diffusion for Consistent 3D Editingcited by 5
  28. 2024Jointly Generating Multi-view Consistent PBR Textures using Collaborative Controlcited by 4
  29. 2025High-fidelity 3D Object Generation from Single Image with RGBN-Volume Gaussian Reconstruction Modelcited by 4
  30. 2025Multi-view Gaze Target Estimationcited by 4
  31. 2025GaMO: Geometry-aware Multi-view Diffusion Outpainting for Sparse-View 3D Reconstructioncited by 4
  32. 2024Sketch-guided Cage-based 3D Gaussian Splatting Deformationcited by 3
  33. 2025ProSplat: Improved Feed-Forward 3D Gaussian Splatting for Wide-Baseline Sparse Viewscited by 3
  34. 2025Elastic3D: Controllable Stereo Video Conversion with Guided Latent Decodingcited by 3
  35. 2026Repurposing 3D Generative Model for Autoregressive Layout Generationcited by 3
  36. 2025MD-ProjTex: Texturing 3D Shapes with Multi-Diffusion Projectioncited by 2
  37. 2025Refining Few-Step Text-to-Multiview Diffusion via Reinforcement Learningcited by 2
  38. 2025AlignCVC: Aligning Cross-View Consistency for Single-Image-to-3D Generationcited by 2

50 most cited of 72 All 72 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025Synthesizing Consistent Novel Views via 3D Epipolar Attention without Re-Training
  2. 2024MVDiff: Scalable and Flexible Multi-View Diffusion for 3D Object Reconstruction from Single-View
  3. 2024MultiDiff: Consistent Novel View Synthesis from a Single Image
  4. 2023Efficient-3DiM: Learning a Generalizable Single-image Novel-view Synthesizer in One Day
  5. 2025DreamComposer++: Empowering Diffusion Models with Multi-View Conditions for 3D Content Generation
  6. 2025CAMEO: Correspondence-Attention Alignment for Multi-View Diffusion Models
  7. 2024Sharp-It: A Multi-view to Multi-view Diffusion Model for 3D Synthesis and Manipulation
  8. 2023MVDream: Multi-view Diffusion for 3D Generation