mlsanityData to 14 Oct 2026

PF-LRM: Pose-Free Large Reconstruction Model for Joint Pose and Shape Prediction

arXiv 2311.12024 ↗First posted 20 Nov 2023cs.CVcited by 165

Not on the map. Why?

Cited by 31 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

We propose a Pose-Free Large Reconstruction Model (PF-LRM) for reconstructing a 3D object from a few unposed images even with little visual overlap, while simultaneously estimating the relative camera poses in ~1.3 seconds on a single A100 GPU. PF-LRM is a highly scalable method utilizing the self-attention blocks to exchange information between 3D object tokens and 2D image tokens; we predict a coarse point cloud for each view, and then use a differentiable Perspective-n-Point (PnP) solver to obtain camera poses. When trained on a huge amount of multi-view posed data of ~1M objects, PF-LRM shows strong cross-dataset generalization ability, and outperforms baseline methods by a large margin in terms of pose prediction accuracy and 3D reconstruction quality on various unseen evaluation datasets. We also demonstrate our model's applicability in downstream text/image-to-3D task with fast feed-forward inference. Our project website is at: https://totoro97.github.io/pf-lrm .

Key facts. Cited by 165 papers in the map (data through 14 Oct 2026). Builds on 61 of its 77 references that are in the map. First posted on arXiv 20 Nov 2023.

Builds on

61 of 77 references.

References · 61

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  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. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,968
  5. 2022Scalable Diffusion Models with Transformerscited by 6,054
  6. 2022Instant Neural Graphics Primitives with a Multiresolution Hash Encodingcited by 3,723
  7. 2022DreamFusion: Text-to-3D using 2D Diffusioncited by 3,025
  8. 2023FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioningcited by 2,598
  9. 2021Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fieldscited by 2,032
  10. 2017Arbitrary Style Transfer in Real-time with Adaptive Instance Normalizationcited by 2,010
  11. 2021NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstructioncited by 1,712
  12. 2020pixelNeRF: Neural Radiance Fields from One or Few Imagescited by 1,584
38 more
  1. 2022Objaverse: A Universe of Annotated 3D Objectscited by 1,540
  2. 2023Zero-1-to-3: Zero-shot One Image to 3D Objectcited by 1,489
  3. 2019SuperGlue: Learning Feature Matching with Graph Neural Networkscited by 1,443
  4. 2017SuperPoint: Self-Supervised Interest Point Detection and Descriptioncited by 1,320
  5. 2019VisualBERT: A Simple and Performant Baseline for Vision and Languagecited by 1,296
  6. 2022Magic3D: High-Resolution Text-to-3D Content Creationcited by 1,290
  7. 2022TensoRF: Tensorial Radiance Fieldscited by 1,280
  8. 2021Efficient Geometry-aware 3D Generative Adversarial Networkscited by 1,205
  9. 2021Volume Rendering of Neural Implicit Surfacescited by 996
  10. 2023MVDream: Multi-view Diffusion for 3D Generationcited by 938
  11. 2020NeRF++: Analyzing and Improving Neural Radiance Fieldscited by 861
  12. 2023LRM: Large Reconstruction Model for Single Image to 3Dcited by 849
  13. 2020Convolutional Occupancy Networkscited by 761
  14. 2021MVSNeRF: Fast Generalizable Radiance Field Reconstruction from Multi-View Stereocited by 740
  15. 2021Depth-supervised NeRF: Fewer Views and Faster Training for Freecited by 735
  16. 2021Common Objects in 3D: Large-Scale Learning and Evaluation of Real-life 3D Category Reconstructioncited by 652
  17. 2023SyncDreamer: Generating Multiview-consistent Images from a Single-view Imagecited by 645
  18. 2021RegNeRF: Regularizing Neural Radiance Fields for View Synthesis from Sparse Inputscited by 615
  19. 2022Google Scanned Objects: A High-Quality Dataset of 3D Scanned Household Itemscited by 612
  20. 2021Ref-NeRF: Structured View-Dependent Appearance for Neural Radiance Fieldscited by 575
  21. 2018From Coarse to Fine: Robust Hierarchical Localization at Large Scalecited by 559
  22. 2023Zero123++: a Single Image to Consistent Multi-view Diffusion Base Modelcited by 521
  23. 2023Instant3D: Fast Text-to-3D with Sparse-View Generation and Large Reconstruction Modelcited by 378
  24. 2021ABO: Dataset and Benchmarks for Real-World 3D Object Understandingcited by 366
  25. 2019D2-Net: A Trainable CNN for Joint Detection and Description of Local Featurescited by 326
  26. 2023OmniObject3D: Large-Vocabulary 3D Object Dataset for Realistic Perception, Reconstruction and Generationcited by 326
  27. 2023FreeNeRF: Improving Few-shot Neural Rendering with Free Frequency Regularizationcited by 311
  28. 2016DSAC - Differentiable RANSAC for Camera Localizationcited by 266
  29. 2023SparseNeRF: Distilling Depth Ranking for Few-shot Novel View Synthesiscited by 247
  30. 2021Scene Representation Transformer: Geometry-Free Novel View Synthesis Through Set-Latent Scene Representationscited by 227
  31. 2023MVImgNet: A Large-scale Dataset of Multi-view Imagescited by 220
  32. 2022SparseFusion: Distilling View-conditioned Diffusion for 3D Reconstructioncited by 216
  33. 2019R2D2: Repeatable and Reliable Detector and Descriptorcited by 207
  34. 2021InfoNeRF: Ray Entropy Minimization for Few-Shot Neural Volume Renderingcited by 182
  35. 2022SparseNeuS: Fast Generalizable Neural Surface Reconstruction from Sparse Viewscited by 177
  36. 2022HuMMan: Multi-Modal 4D Human Dataset for Versatile Sensing and Modelingcited by 117
  37. 2023RelPose++: Recovering 6D Poses from Sparse-view Observationscited by 87
  38. 2023EPro-PnP: Generalized End-to-End Probabilistic Perspective-n-Points for Monocular Object Pose Estimationcited by 79

50 most cited of 61 All 61 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 2; 2024: 79; 2025: 59; 2026: 25.

165 papers.

Sort:

By area:

  1. 2025VGGT: Visual Geometry Grounded Transformercited by 1,539
  2. 2024LGM: Large Multi-View Gaussian Model for High-Resolution 3D Content Creationcited by 739
  3. 2024InstantMesh: Efficient 3D Mesh Generation from a Single Image with Sparse-view Large Reconstruction Modelscited by 433
  4. 2024GS-LRM: Large Reconstruction Model for 3D Gaussian Splattingcited by 349
  5. 2025MapAnything: Universal Feed-Forward Metric 3D Reconstructioncited by 314
  6. 2024GRM: Large Gaussian Reconstruction Model for Efficient 3D Reconstruction and Generationcited by 301
  7. 2023Triplane Meets Gaussian Splatting: Fast and Generalizable Single-View 3D Reconstruction with Transformerscited by 258
  8. 2025AnySplat: Feed-forward 3D Gaussian Splatting from Unconstrained Viewscited by 243
  9. 2024TripoSR: Fast 3D Object Reconstruction from a Single Imagecited by 232
  10. 2024CRM: Single Image to 3D Textured Mesh with Convolutional Reconstruction Modelcited by 215
  11. 2025TripoSG: High-Fidelity 3D Shape Synthesis using Large-Scale Rectified Flow Modelscited by 213
  12. 2024Direct3D: Scalable Image-to-3D Generation via 3D Latent Diffusion Transformercited by 210
38 more
  1. 2024LVSM: A Large View Synthesis Model with Minimal 3D Inductive Biascited by 189
  2. 2025FLARE: Feed-forward Geometry, Appearance and Camera Estimation from Uncalibrated Sparse Viewscited by 180
  3. 2024SV4D: Dynamic 3D Content Generation with Multi-Frame and Multi-View Consistencycited by 144
  4. 2024MeshLRM: Large Reconstruction Model for High-Quality Meshescited by 139
  5. 2024Cameras as Rays: Pose Estimation via Ray Diffusioncited by 131
  6. 2024Long-LRM: Long-sequence Large Reconstruction Model for Wide-coverage Gaussian Splatscited by 114
  7. 2024MeshXL: Neural Coordinate Field for Generative 3D Foundation Modelscited by 103
  8. 20243DTopia-XL: Scaling High-quality 3D Asset Generation via Primitive Diffusioncited by 101
  9. 2024GaussianFlow: Splatting Gaussian Dynamics for 4D Content Creationcited by 100
  10. 2024Era3D: High-Resolution Multiview Diffusion using Efficient Row-wise Attentioncited by 100
  11. 2024Dora: Sampling and Benchmarking for 3D Shape Variational Auto-Encoderscited by 98
  12. 2024MVDiffusion++: A Dense High-resolution Multi-view Diffusion Model for Single or Sparse-view 3D Object Reconstructioncited by 95
  13. 2024LLaMA-Mesh: Unifying 3D Mesh Generation with Language Modelscited by 91
  14. 2024EdgeRunner: Auto-regressive Auto-encoder for Artistic Mesh Generationcited by 87
  15. 2025RayZer: A Self-supervised Large View Synthesis Modelcited by 77
  16. 2025WorldMirror: Universal 3D World Reconstruction with Any-Prior Promptingcited by 67
  17. 2024SF3D: Stable Fast 3D Mesh Reconstruction with UV-unwrapping and Illumination Disentanglementcited by 65
  18. 2024You See it, You Got it: Learning 3D Creation on Pose-Free Videos at Scalecited by 65
  19. 2025DeepMesh: Auto-Regressive Artist-mesh Creation with Reinforcement Learningcited by 65
  20. 2024GaussianObject: High-Quality 3D Object Reconstruction from Four Views with Gaussian Splattingcited by 59
  21. 2024MegaScenes: Scene-Level View Synthesis at Scalecited by 58
  22. 2024VFusion3D: Learning Scalable 3D Generative Models from Video Diffusion Modelscited by 56
  23. 2024Gamba: Marry Gaussian Splatting with Mamba for single view 3D reconstructioncited by 56
  24. 2024LaRa: Efficient Large-Baseline Radiance Fieldscited by 55
  25. 2024MeshFormer: High-Quality Mesh Generation with 3D-Guided Reconstruction Modelcited by 55
  26. 2025Avat3r: Large Animatable Gaussian Reconstruction Model for High-fidelity 3D Head Avatarscited by 53
  27. 2025SV4D 2.0: Enhancing Spatio-Temporal Consistency in Multi-View Video Diffusion for High-Quality 4D Generationcited by 48
  28. 2024TEXGen: a Generative Diffusion Model for Mesh Texturescited by 46
  29. 2024VideoMV: Consistent Multi-View Generation Based on Large Video Generative Modelcited by 45
  30. 2025Ultra3D: Efficient and High-Fidelity 3D Generation with Part Attentioncited by 44
  31. 2024SelfSplat: Pose-Free and 3D Prior-Free Generalizable 3D Gaussian Splattingcited by 42
  32. 2024FreeSplatter: Pose-free Gaussian Splatting for Sparse-view 3D Reconstructioncited by 41
  33. 2025Matrix3D: Large Photogrammetry Model All-in-Onecited by 40
  34. 2025No Pose at All: Self-Supervised Pose-Free 3D Gaussian Splatting from Sparse Viewscited by 40
  35. 2025ShapeLLM-Omni: A Native Multimodal LLM for 3D Generation and Understandingcited by 38
  36. 2024TIPS: Text-Image Pretraining with Spatial awarenesscited by 37
  37. 2026LingBot-Map: Geometric Context Transformer for Streaming 3D Reconstructioncited by 37
  38. 2024GeoLRM: Geometry-Aware Large Reconstruction Model for High-Quality 3D Gaussian Generationcited by 36

50 most cited of 165 All 165 (JSON)