mlsanityData to 13 Oct 2026

latentSplat: Autoencoding Variational Gaussians for Fast Generalizable 3D Reconstruction

arXiv 2403.16292 ↗First posted 24 Mar 2024cs.CVcited by 116

Not on the map. Why?

Cited by 21 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

We present latentSplat, a method to predict semantic Gaussians in a 3D latent space that can be splatted and decoded by a light-weight generative 2D architecture. Existing methods for generalizable 3D reconstruction either do not scale to large scenes and resolutions, or are limited to interpolation of close input views. latentSplat combines the strengths of regression-based and generative approaches while being trained purely on readily available real video data. The core of our method are variational 3D Gaussians, a representation that efficiently encodes varying uncertainty within a latent space consisting of 3D feature Gaussians. From these Gaussians, specific instances can be sampled and rendered via efficient splatting and a fast, generative decoder. We show that latentSplat outperforms previous works in reconstruction quality and generalization, while being fast and scalable to high-resolution data.

Key facts. Cited by 116 papers in the map (data through 13 Oct 2026). Builds on 74 of its 82 references that are in the map. First posted on arXiv 24 Mar 2024.

Builds on

74 of 82 references.

References · 74

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,952
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,777
  3. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,082
  4. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,213
  5. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,915
  6. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,564
  7. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,339
  8. 20233D Gaussian Splatting for Real-Time Radiance Field Renderingcited by 6,170
  9. 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,516
  10. 2023SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesiscited by 4,323
  11. 2021GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Modelscited by 3,605
  12. 2019DeepSDF: Learning Continuous Signed Distance Functions for Shape Representationcited by 2,686
38 more
  1. 2020Implicit Neural Representations with Periodic Activation Functionscited by 2,225
  2. 2018Occupancy Networks: Learning 3D Reconstruction in Function Spacecited by 2,039
  3. 2021NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstructioncited by 1,706
  4. 2020pixelNeRF: Neural Radiance Fields from One or Few Imagescited by 1,581
  5. 2023Zero-1-to-3: Zero-shot One Image to 3D Objectcited by 1,486
  6. 2018Demystifying MMD GANscited by 1,277
  7. 2021Efficient Geometry-aware 3D Generative Adversarial Networkscited by 1,205
  8. 20234D Gaussian Splatting for Real-Time Dynamic Scene Renderingcited by 1,150
  9. 2021Volume Rendering of Neural Implicit Surfacescited by 992
  10. 2019Scene Representation Networks: Continuous 3D-Structure-Aware Neural Scene Representationscited by 864
  11. 2020Image Quality Assessment: Unifying Structure and Texture Similaritycited by 800
  12. 2021MVSNeRF: Fast Generalizable Radiance Field Reconstruction from Multi-View Stereocited by 738
  13. 2020pi-GAN: Periodic Implicit Generative Adversarial Networks for 3D-Aware Image Synthesiscited by 708
  14. 2019Differentiable Volumetric Rendering: Learning Implicit 3D Representations without 3D Supervisioncited by 699
  15. 2020GRAF: Generative Radiance Fields for 3D-Aware Image Synthesiscited by 689
  16. 2023pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstructioncited by 673
  17. 2021Common Objects in 3D: Large-Scale Learning and Evaluation of Real-life 3D Category Reconstructioncited by 648
  18. 2023SuGaR: Surface-Aligned Gaussian Splatting for Efficient 3D Mesh Reconstruction and High-Quality Mesh Renderingcited by 613
  19. 2023One-2-3-45: Any Single Image to 3D Mesh in 45 Seconds without Per-Shape Optimizationcited by 557
  20. 2019SynSin: End-to-end View Synthesis from a Single Imagecited by 364
  21. 2023Splatter Image: Ultra-Fast Single-View 3D Reconstructioncited by 362
  22. 2021Light Field Networks: Neural Scene Representations with Single-Evaluation Renderingcited by 332
  23. 2019Neural Point-Based Graphicscited by 326
  24. 2020Deep Local Shapes: Learning Local SDF Priors for Detailed 3D Reconstructioncited by 321
  25. 2023ReconFusion: 3D Reconstruction with Diffusion Priorscited by 317
  26. 2022Novel View Synthesis with Diffusion Modelscited by 300
  27. 20223D Neural Field Generation using Triplane Diffusioncited by 274
  28. 2023Generative Novel View Synthesis with 3D-Aware Diffusion Modelscited by 273
  29. 2020Single-View View Synthesis with Multiplane Imagescited by 263
  30. 2020Free View Synthesiscited by 231
  31. 2022SparseFusion: Distilling View-conditioned Diffusion for 3D Reconstructioncited by 216
  32. 2022DiffRF: Rendering-Guided 3D Radiance Field Diffusioncited by 204
  33. 2021GeoNeRF: Generalizing NeRF with Geometry Priorscited by 195
  34. 2021Neural Rays for Occlusion-aware Image-based Renderingcited by 192
  35. 2019Pushing the Boundaries of View Extrapolation with Multiplane Imagescited by 191
  36. 2021Stereo Radiance Fields (SRF): Learning View Synthesis for Sparse Views of Novel Scenescited by 190
  37. 2021CodeNeRF: Disentangled Neural Radiance Fields for Object Categoriescited by 183
  38. 2023Single-Stage Diffusion NeRF: A Unified Approach to 3D Generation and Reconstructioncited by 180

50 most cited of 74 All 74 (JSON)

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2024: 37; 2025: 43; 2026: 36.

116 papers.

Sort:

By area:

  1. 2024DepthSplat: Connecting Gaussian Splatting and Depthcited by 277
  2. 2024Splatt3R: Zero-shot Gaussian Splatting from Uncalibrated Image Pairscited by 174
  3. 2024ReconX: Reconstruct Any Scene from Sparse Views with Video Diffusion Modelcited by 141
  4. 2024MV-DUSt3R+: Single-Stage Scene Reconstruction from Sparse Views In 2 Secondscited by 137
  5. 2024Flash3D: Feed-Forward Generalisable 3D Scene Reconstruction from a Single Imagecited by 114
  6. 2024MVSplat360: Feed-Forward 360 Scene Synthesis from Sparse Viewscited by 112
  7. 2025MEt3R: Measuring Multi-View Consistency in Generated Imagescited by 92
  8. 2024FreeSplat: Generalizable 3D Gaussian Splatting Towards Free-View Synthesis of Indoor Scenescited by 87
  9. 2024TranSplat: Generalizable 3D Gaussian Splatting from Sparse Multi-View Images with Transformerscited by 72
  10. 2024FlowMap: High-Quality Camera Poses, Intrinsics, and Depth via Gradient Descentcited by 61
  11. 2025Bolt3D: Generating 3D Scenes in Secondscited by 57
  12. 2024HiSplat: Hierarchical 3D Gaussian Splatting for Generalizable Sparse-View Reconstructioncited by 55
38 more
  1. 2024STORM: Spatio-Temporal Reconstruction Model for Large-Scale Outdoor Scenescited by 53
  2. 2025Amodal3R: Amodal 3D Reconstruction from Occluded 2D Imagescited by 52
  3. 2024GGHead: Fast and Generalizable 3D Gaussian Headscited by 51
  4. 2024PhysMotion: Physics-Grounded Dynamics From a Single Imagecited by 45
  5. 2024L3DG: Latent 3D Gaussian Diffusioncited by 42
  6. 2024FreeSplatter: Pose-free Gaussian Splatting for Sparse-view 3D Reconstructioncited by 40
  7. 2024SeaSplat: Representing Underwater Scenes with 3D Gaussian Splatting and a Physically Grounded Image Formation Modelcited by 35
  8. 2024Prometheus: 3D-Aware Latent Diffusion Models for Feed-Forward Text-to-3D Scene Generationcited by 33
  9. 2025StarGen: A Spatiotemporal Autoregression Framework with Video Diffusion Model for Scalable and Controllable Scene Generationcited by 26
  10. 2026Gen3R: 3D Scene Generation Meets Feed-Forward Reconstructioncited by 26
  11. 2024LM-Gaussian: Boost Sparse-view 3D Gaussian Splatting with Large Model Priorscited by 25
  12. 2025Advances in Feed-Forward 3D Reconstruction and View Synthesis: A Surveycited by 25
  13. 2024PanSplat: 4K Panorama Synthesis with Feed-Forward Gaussian Splattingcited by 24
  14. 2025DSO: Aligning 3D Generators with Simulation Feedback for Physical Soundnesscited by 24
  15. 2024MegaSynth: Scaling Up 3D Scene Reconstruction with Synthesized Datacited by 21
  16. 2025CamCtrl3D: Single-Image Scene Exploration with Precise 3D Camera Controlcited by 21
  17. 2025EVolSplat: Efficient Volume-based Gaussian Splatting for Urban View Synthesiscited by 21
  18. 2024MVGS: Multi-view Regulated Gaussian Splatting for Novel View Synthesiscited by 20
  19. 2025Generative Gaussian Splatting: Generating 3D Scenes with Video Diffusion Priorscited by 19
  20. 2025A Survey on 3D Gaussian Splatting Applications: Segmentation, Editing, and Generationcited by 19
  21. 2024Epipolar-Free 3D Gaussian Splatting for Generalizable Novel View Synthesiscited by 18
  22. 2024Generative Densification: Learning to Densify Gaussians for High-Fidelity Generalizable 3D Reconstructioncited by 18
  23. 2025iLRM: An Iterative Large 3D Reconstruction Modelcited by 18
  24. 2024SplatFormer: Point Transformer for Robust 3D Gaussian Splattingcited by 15
  25. 2025MonoSplat: Generalizable 3D Gaussian Splatting from Monocular Depth Foundation Modelscited by 15
  26. 2024GSemSplat: Generalizable Semantic 3D Gaussian Splatting from Uncalibrated Image Pairscited by 14
  27. 2024Extrapolated Urban View Synthesis Benchmarkcited by 13
  28. 2025FreeSplat++: Generalizable 3D Gaussian Splatting for Efficient Indoor Scene Reconstructioncited by 13
  29. 2026TokenSplat: Token-aligned 3D Gaussian Splatting for Feed-forward Pose-free Reconstructioncited by 13
  30. 2025LongSplat: Online Generalizable 3D Gaussian Splatting from Long Sequence Imagescited by 12
  31. 2024Sampling 3D Gaussian Scenes in Seconds with Latent Diffusion Modelscited by 10
  32. 2024OmniSplat: Taming Feed-Forward 3D Gaussian Splatting for Omnidirectional Images with Editable Capabilitiescited by 10
  33. 2026One2Scene: Geometric Consistent Explorable 3D Scene Generation from a Single Imagecited by 10
  34. 2024Mipmap-GS: Let Gaussians Deform with Scale-specific Mipmap for Anti-aliasing Renderingcited by 9
  35. 2024Spurfies: Sparse Surface Reconstruction using Local Geometry Priorscited by 8
  36. 2024Gaussian Scenes: Pose-Free Sparse-View Scene Reconstruction using Depth-Enhanced Diffusion Priorscited by 8
  37. 2024Sp2360: Sparse-view 360 Scene Reconstruction using Cascaded 2D Diffusion Priorscited by 7
  38. 2025Seeing World Dynamics in a Nutshellcited by 7

50 most cited of 116 All 116 (JSON)