mlsanityData to 14 Oct 2026

MVSplat360: Feed-Forward 360 Scene Synthesis from Sparse Views

arXiv 2411.04924 ↗First posted 7 Nov 2024cs.CVcited by 116

Not on the map. Why?

Cited by 25 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

We introduce MVSplat360, a feed-forward approach for 360deg novel view synthesis (NVS) of diverse real-world scenes, using only sparse observations. This setting is inherently ill-posed due to minimal overlap among input views and insufficient visual information provided, making it challenging for conventional methods to achieve high-quality results. Our MVSplat360 addresses this by effectively combining geometry-aware 3D reconstruction with temporally consistent video generation. Specifically, it refactors a feed-forward 3D Gaussian Splatting (3DGS) model to render features directly into the latent space of a pre-trained Stable Video Diffusion (SVD) model, where these features then act as pose and visual cues to guide the denoising process and produce photorealistic 3D-consistent views. Our model is end-to-end trainable and supports rendering arbitrary views with as few as 5 sparse input views. To evaluate MVSplat360's performance, we introduce a new benchmark using the challenging DL3DV-10K dataset, where MVSplat360 achieves superior visual quality compared to state-of-the-art methods on wide-sweeping or even 360deg NVS tasks. Experiments on the existing benchmark RealEstate10K also confirm the effectiveness of our model. The video results are available on our project page: https://donydchen.github.io/mvsplat360.

Key facts. Cited by 116 papers in the map (data through 14 Oct 2026). Builds on 66 of its 77 references that are in the map. First posted on arXiv 7 Nov 2024.

Builds on

66 of 77 references.

References · 66

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  3. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,177
  4. 20233D Gaussian Splatting for Real-Time Radiance Field Renderingcited by 6,226
  5. 2020Taming Transformers for High-Resolution Image Synthesiscited by 3,253
  6. 2023Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasetscited by 2,581
  7. 2021Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fieldscited by 2,032
  8. 2021Mip-NeRF: A Multiscale Representation for Anti-Aliasing Neural Radiance Fieldscited by 1,964
  9. 2022Imagen Video: High Definition Video Generation with Diffusion Modelscited by 1,778
  10. 2022Make-A-Video: Text-to-Video Generation without Text-Video Datacited by 1,742
  11. 2020pixelNeRF: Neural Radiance Fields from One or Few Imagescited by 1,584
  12. 2023Zero-1-to-3: Zero-shot One Image to 3D Objectcited by 1,489
38 more
  1. 2022MaskGIT: Masked Generative Image Transformercited by 1,099
  2. 2019Scene Representation Networks: Continuous 3D-Structure-Aware Neural Scene Representationscited by 864
  3. 2020Image Quality Assessment: Unifying Structure and Texture Similaritycited by 811
  4. 2021MVSNeRF: Fast Generalizable Radiance Field Reconstruction from Multi-View Stereocited by 740
  5. 2024LGM: Large Multi-View Gaussian Model for High-Resolution 3D Content Creationcited by 739
  6. 2021Depth-supervised NeRF: Fewer Views and Faster Training for Freecited by 735
  7. 2021IBRNet: Learning Multi-View Image-Based Renderingcited by 703
  8. 2023pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstructioncited by 684
  9. 2018Pixel2Mesh: Generating 3D Mesh Models from Single RGB Imagescited by 651
  10. 2021RegNeRF: Regularizing Neural Radiance Fields for View Synthesis from Sparse Inputscited by 615
  11. 2023MotionCtrl: A Unified and Flexible Motion Controller for Video Generationcited by 580
  12. 2024MVSplat: Efficient 3D Gaussian Splatting from Sparse Multi-View Imagescited by 566
  13. 2023One-2-3-45: Any Single Image to 3D Mesh in 45 Seconds without Per-Shape Optimizationcited by 560
  14. 2023DL3DV-10K: A Large-Scale Scene Dataset for Deep Learning-based 3D Visioncited by 537
  15. 2022MonoSDF: Exploring Monocular Geometric Cues for Neural Implicit Surface Reconstructioncited by 470
  16. 2024CAT3D: Create Anything in 3D with Multi-View Diffusion Modelscited by 432
  17. 2024ViewCrafter: Taming Video Diffusion Models for High-fidelity Novel View Synthesiscited by 403
  18. 2023Splatter Image: Ultra-Fast Single-View 3D Reconstructioncited by 364
  19. 2024SV3D: Novel Multi-view Synthesis and 3D Generation from a Single Image using Latent Video Diffusioncited by 364
  20. 2024GS-LRM: Large Reconstruction Model for 3D Gaussian Splattingcited by 349
  21. 2021Light Field Networks: Neural Scene Representations with Single-Evaluation Renderingcited by 332
  22. 2023ReconFusion: 3D Reconstruction with Diffusion Priorscited by 318
  23. 2018Learning Category-Specific Mesh Reconstruction from Image Collectionscited by 317
  24. 2024GRM: Large Gaussian Reconstruction Model for Efficient 3D Reconstruction and Generationcited by 301
  25. 2024DepthSplat: Connecting Gaussian Splatting and Depthcited by 286
  26. 2023One-2-3-45++: Fast Single Image to 3D Objects with Consistent Multi-View Generation and 3D Diffusioncited by 283
  27. 2023Generative Novel View Synthesis with 3D-Aware Diffusion Modelscited by 274
  28. 2021Scene Representation Transformer: Geometry-Free Novel View Synthesis Through Set-Latent Scene Representationscited by 227
  29. 2023Visual Geometry Grounded Deep Structure From Motioncited by 227
  30. 2021GeoNeRF: Generalizing NeRF with Geometry Priorscited by 196
  31. 2022SPARF: Neural Radiance Fields from Sparse and Noisy Posescited by 195
  32. 2024LVSM: A Large View Synthesis Model with Minimal 3D Inductive Biascited by 189
  33. 2024Splatt3R: Zero-shot Gaussian Splatting from Uncalibrated Image Pairscited by 178
  34. 2024ReconX: Reconstruct Any Scene from Sparse Views with Video Diffusion Modelcited by 143
  35. 2022MoVQ: Modulating Quantized Vectors for High-Fidelity Image Generationcited by 125
  36. 2022Generalizable Patch-Based Neural Renderingcited by 122
  37. 2024latentSplat: Autoencoding Variational Gaussians for Fast Generalizable 3D Reconstructioncited by 118
  38. 2023Consistent View Synthesis with Pose-Guided Diffusion Modelscited by 117

50 most cited of 66 All 66 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 7; 2025: 54; 2026: 55.

116 papers.

Sort:

By area:

  1. 2025AnySplat: Feed-forward 3D Gaussian Splatting from Unconstrained Viewscited by 243
  2. 2025VMem: Consistent Interactive Video Scene Generation with Surfel-Indexed View Memorycited by 109
  3. 2025Cameras as Relative Positional Encodingcited by 105
  4. 2025Geo4D: Leveraging Video Generators for Geometric 4D Scene Reconstructioncited by 67
  5. 2025Bolt3D: Generating 3D Scenes in Secondscited by 59
  6. 2025Amodal3R: Amodal 3D Reconstruction from Occluded 2D Imagescited by 54
  7. 2024StreetCrafter: Street View Synthesis with Controllable Video Diffusion Modelscited by 49
  8. 2025VolSplat: Rethinking Feed-Forward 3D Gaussian Splatting with Voxel-Aligned Predictioncited by 48
  9. 2024Prometheus: 3D-Aware Latent Diffusion Models for Feed-Forward Text-to-3D Scene Generationcited by 34
  10. 2025WonderTurbo: Generating Interactive 3D World in 0.72 Secondscited by 32
  11. 2025ZPressor: Bottleneck-Aware Compression for Scalable Feed-Forward 3DGScited by 32
  12. 2026Gen3R: 3D Scene Generation Meets Feed-Forward Reconstructioncited by 31
38 more
  1. 2025FlowR: Flowing from Sparse to Dense 3D Reconstructionscited by 28
  2. 2025VideoScene: Distilling Video Diffusion Model to Generate 3D Scenes in One Stepcited by 27
  3. 2025Advances in Feed-Forward 3D Reconstruction and View Synthesis: A Surveycited by 27
  4. 2025DSO: Aligning 3D Generators with Simulation Feedback for Physical Soundnesscited by 25
  5. 2024Splatter-360: Generalizable 360circ Gaussian Splatting for Wide-baseline Panoramic Imagescited by 24
  6. 2024MegaSynth: Scaling Up 3D Scene Reconstruction with Synthesized Datacited by 21
  7. 2025iLRM: An Iterative Large 3D Reconstruction Modelcited by 19
  8. 2025A Survey on 3D Gaussian Splatting Applications: Segmentation, Editing, and Generationcited by 19
  9. 2024Generative Densification: Learning to Densify Gaussians for High-Fidelity Generalizable 3D Reconstructioncited by 18
  10. 2024SLGaussian: Fast Language Gaussian Splatting in Sparse Viewscited by 18
  11. 2025FreeSplat++: Generalizable 3D Gaussian Splatting for Efficient Indoor Scene Reconstructioncited by 15
  12. 2025LightSwitch: Multi-view Relighting with Material-guided Diffusioncited by 15
  13. 2025ReSplat: Learning Recurrent Gaussian Splattingcited by 15
  14. 2025LL-Gaussian: Low-Light Scene Reconstruction and Enhancement via Gaussian Splatting for Novel View Synthesiscited by 14
  15. 20254Real-Video-V2: Fused View-Time Attention and Feedforward Reconstruction for 4D Scene Generationcited by 14
  16. 2025Quantifying and Alleviating Co-Adaptation in Sparse-View 3D Gaussian Splattingcited by 14
  17. 2025Feed-Forward SceneDINO for Unsupervised Semantic Scene Completioncited by 13
  18. 2025LongSplat: Online Generalizable 3D Gaussian Splatting from Long Sequence Imagescited by 13
  19. 2025Revisiting Depth Representations for Feed-Forward 3D Gaussian Splattingcited by 12
  20. 2025Off The Grid: Detection of Primitives for Feed-Forward 3D Gaussian Splattingcited by 11
  21. 2026One2Scene: Geometric Consistent Explorable 3D Scene Generation from a Single Imagecited by 11
  22. 2025Ov3R: Open-Vocabulary Semantic 3D Reconstruction from RGB Videoscited by 10
  23. 2026SceneTok: A Compressed, Diffusable Token Space for 3D Scenescited by 10
  24. 2026LagerNVS: Latent Geometry for Fully Neural Real-time Novel View Synthesiscited by 9
  25. 2025VideoFrom3D: 3D Scene Video Generation via Complementary Image and Video Diffusion Modelscited by 8
  26. 2025TinySplat: Feedforward Approach for Generating Compact 3D Scene Representationcited by 7
  27. 2025WonderFree: Enhancing Novel View Quality and Cross-View Consistency for 3D Scene Explorationcited by 7
  28. 2026NOVA3R: Non-pixel-aligned Visual Transformer for Amodal 3D Reconstructioncited by 7
  29. 2026GaussFusion: Improving 3D Reconstruction in the Wild with A Geometry-Informed Video Generatorcited by 7
  30. 2026SparseSplat: Towards Applicable Feed-Forward 3D Gaussian Splatting with Pixel-Unaligned Predictioncited by 7
  31. 2024Motion-Aware Animatable Gaussian Avatars Deblurringcited by 6
  32. 2025TextSplat: Text-Guided Semantic Fusion for Generalizable Gaussian Splattingcited by 6
  33. 2025EarthCrafter: Scalable 3D Earth Generation via Dual-Sparse Latent Diffusioncited by 6
  34. 2025CausNVS: Autoregressive Multi-view Diffusion for Flexible 3D Novel View Synthesiscited by 6
  35. 2026One-Shot Refiner: Boosting Feed-forward Novel View Synthesis via One-Step Diffusioncited by 6
  36. 2025HouseTour: A Virtual Real Estate A(I)gentcited by 5
  37. 2025ProSplat: Improved Feed-Forward 3D Gaussian Splatting for Wide-Baseline Sparse Viewscited by 4
  38. 2025FRIDU: Functional Map Refinement with Guided Image Diffusioncited by 4

50 most cited of 116 All 116 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2024Sp2360: Sparse-view 360 Scene Reconstruction using Cascaded 2D Diffusion Priors
  2. 2025Taming Video Diffusion Prior with Scene-Grounding Guidance for 3D Gaussian Splatting from Sparse Inputs
  3. 2026GaussVid: Sparse-View Gaussian Splatting with 3D-Aware Video Diffusion Priors
  4. 2025Improving Novel view synthesis of 360^circ Scenes in Extremely Sparse Views by Jointly Training Hemisphere Sampled Synthetic Images
  5. 2026ArtiFixer: Enhancing and Extending 3D Reconstruction with Auto-Regressive Diffusion Models
  6. 2026DensiTok: Making Feed-Forward 3D Gaussian Splatting See More Views Than It Is Given
  7. 2026BetterScene: 3D Scene Synthesis with Representation-Aligned Generative Model
  8. 2026VidSplat: Gaussian Splatting Reconstruction with Geometry-Guided Video Diffusion Priors