mlsanityData to 7 Oct 2026

Amodal3R: Amodal 3D Reconstruction from Occluded 2D Images

arXiv 2503.13439 ↗First posted 17 Mar 2025cs.CVcited by 52

Not on the map. Why?

Abstract (arXiv)

Most image-based 3D object reconstructors assume that objects are fully visible, ignoring occlusions that commonly occur in real-world scenarios. In this paper, we introduce Amodal3R, a conditional 3D generative model designed to reconstruct 3D objects from partial observations. We start from a "foundation" 3D generative model and extend it to recover plausible 3D geometry and appearance from occluded objects. We introduce a mask-weighted multi-head cross-attention mechanism followed by an occlusion-aware attention layer that explicitly leverages occlusion priors to guide the reconstruction process. We demonstrate that, by training solely on synthetic data, Amodal3R learns to recover full 3D objects even in the presence of occlusions in real scenes. It substantially outperforms existing methods that independently perform 2D amodal completion followed by 3D reconstruction, thereby establishing a new benchmark for occlusion-aware 3D reconstruction.

Key facts. Cited by 52 papers in the map (data through 7 Oct 2026). Builds on 89 of its 101 references that are in the map. First posted on arXiv 17 Mar 2025.

Builds on

89 of 101 references.

References · 89

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  4. 2023Segment Anythingcited by 8,158
  5. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,764
  6. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  7. 20233D Gaussian Splatting for Real-Time Radiance Field Renderingcited by 6,141
  8. 2022Classifier-Free Diffusion Guidancecited by 5,717
  9. 2022Generative Adversarial Networkscited by 5,603
  10. 2022Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flowcited by 3,550
  11. 2023Zero-1-to-3: Zero-shot One Image to 3D Objectcited by 1,483
  12. 2016Context Encoders: Feature Learning by Inpaintingcited by 1,424
38 more
  1. 2018Demystifying MMD GANscited by 1,268
  2. 2021Efficient Geometry-aware 3D Generative Adversarial Networkscited by 1,204
  3. 2023MVDream: Multi-view Diffusion for 3D Generationcited by 928
  4. 2019The Replica Dataset: A Digital Replica of Indoor Spacescited by 885
  5. 2023LRM: Large Reconstruction Model for Single Image to 3Dcited by 834
  6. 2020GIRAFFE: Representing Scenes as Compositional Generative Neural Feature Fieldscited by 784
  7. 2023Objaverse-XL: A Universe of 10M+ 3D Objectscited by 767
  8. 2024Structured 3D Latents for Scalable and Versatile 3D Generationcited by 742
  9. 2021Diffusion Probabilistic Models for 3D Point Cloud Generationcited by 734
  10. 2020pi-GAN: Periodic Implicit Generative Adversarial Networks for 3D-Aware Image Synthesiscited by 708
  11. 2022Point-E: A System for Generating 3D Point Clouds from Complex Promptscited by 694
  12. 2020GRAF: Generative Radiance Fields for 3D-Aware Image Synthesiscited by 688
  13. 2023Wonder3D: Single Image to 3D using Cross-Domain Diffusioncited by 682
  14. 2021DiffusionCLIP: Text-Guided Diffusion Models for Robust Image Manipulationcited by 640
  15. 2022Google Scanned Objects: A High-Quality Dataset of 3D Scanned Household Itemscited by 601
  16. 2022LION: Latent Point Diffusion Models for 3D Shape Generationcited by 552
  17. 20213D Shape Generation and Completion through Point-Voxel Diffusioncited by 516
  18. 2023Zero123++: a Single Image to Consistent Multi-view Diffusion Base Modelcited by 516
  19. 2018Image Inpainting for Irregular Holes Using Partial Convolutionscited by 505
  20. 2022GET3D: A Generative Model of High Quality 3D Textured Shapes Learned from Imagescited by 446
  21. 2024CLAY: A Controllable Large-scale Generative Model for Creating High-quality 3D Assetscited by 444
  22. 2023Shap-E: Generating Conditional 3D Implicit Functionscited by 379
  23. 2024SV3D: Novel Multi-view Synthesis and 3D Generation from a Single Image using Latent Video Diffusioncited by 364
  24. 2021ABO: Dataset and Benchmarks for Real-World 3D Object Understandingcited by 360
  25. 2023Splatter Image: Ultra-Fast Single-View 3D Reconstructioncited by 359
  26. 20203D-FUTURE: 3D Furniture shape with TextUREcited by 306
  27. 2022Novel View Synthesis with Diffusion Modelscited by 299
  28. 2022SDFusion: Multimodal 3D Shape Completion, Reconstruction, and Generationcited by 287
  29. 20223D Neural Field Generation using Triplane Diffusioncited by 274
  30. 2020PolyGen: An Autoregressive Generative Model of 3D Meshescited by 267
  31. 2023Triplane Meets Gaussian Splatting: Fast and Generalizable Single-View 3D Reconstruction with Transformerscited by 257
  32. 2023MeshGPT: Generating Triangle Meshes with Decoder-Only Transformerscited by 254
  33. 2024Direct3D: Scalable Image-to-3D Generation via 3D Latent Diffusion Transformercited by 210
  34. 2022DiffRF: Rendering-Guided 3D Radiance Field Diffusioncited by 204
  35. 20233DGen: Triplane Latent Diffusion for Textured Mesh Generationcited by 181
  36. 2019Pluralistic Image Completioncited by 162
  37. 2022RenderDiffusion: Image Diffusion for 3D Reconstruction, Inpainting and Generationcited by 162
  38. 2023Habitat Synthetic Scenes Dataset (HSSD-200): An Analysis of 3D Scene Scale and Realism Tradeoffs for ObjectGoal Navigationcited by 162

50 most cited of 89 All 89 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 18; 2026: 34.

52 papers.

Sort:

By area:

  1. 2025SAM 3D: 3Dfy Anything in Imagescited by 230
  2. 2025VoxHammer: Training-Free Precise and Coherent 3D Editing in Native 3D Spacecited by 40
  3. 2026ShapeR: Robust Conditional 3D Shape Generation from Casual Capturescited by 19
  4. 2025R3D2: Realistic 3D Asset Insertion via Diffusion for Autonomous Driving Simulationcited by 17
  5. 2025GSWorld: Closed-Loop Photo-Realistic Simulation Suite for Robotic Manipulationcited by 16
  6. 2025SceneMaker: Open-set 3D Scene Generation with Decoupled De-occlusion and Pose Estimation Modelcited by 15
  7. 2025Orientation Matters: Making 3D Generative Models Orientation-Alignedcited by 13
  8. 2025HoloScene: Simulation-Ready Interactive 3D Worlds from a Single Videocited by 13
  9. 20253D-RE-GEN: 3D Reconstruction of Indoor Scenes with a Generative Frameworkcited by 10
  10. 2025SpaceControl: Introducing Test-Time Spatial Control to 3D Generative Modelingcited by 9
  11. 2025Video Perception Models for 3D Scene Synthesiscited by 7
  12. 2026NOVA3R: Non-pixel-aligned Visual Transformer for Amodal 3D Reconstructioncited by 5
38 more
  1. 2025FRIDU: Functional Map Refinement with Guided Image Diffusioncited by 4
  2. 2025GENA3D: Generative Amodal 3D Modeling by Bridging 2D Priors and 3D Coherencecited by 4
  3. 2026SeeThrough3D: Occlusion Aware 3D Control in Text-to-Image Generationcited by 4
  4. 2026Points-to-3D: Structure-Aware 3D Generation with Point Cloud Priorscited by 4
  5. 2026Syn4D: A Multiview Synthetic 4D Datasetcited by 4
  6. 2025Fast3Dcache: Training-free 3D Geometry Synthesis Accelerationcited by 3
  7. 2026MotionCrafter: Dense Geometry and Motion Reconstruction with a 4D VAEcited by 3
  8. 2025SynergyAmodal: Deocclude Anything with Text Controlcited by 2
  9. 2026EGM: Efficient Visual Grounding Language Modelscited by 2
  10. 2026Axolotl3D: a Unified Framework for Faithful 3D Shape Completioncited by 2
  11. 2026Lucida: Parse, Generate, and Place for Composable Real-to-Sim Scene Modelingcited by 2
  12. 2026TourPhysics: Bringing Physics to World Models for Exploration and Manipulation from a Single Imagecited by 2
  13. 2026WorldSculpt: Generating Compositional Worlds from Grounded Videoscited by 2
  14. 2025NeoWorld: Neural Simulation of Explorable Virtual Worlds via Progressive 3D Unfoldingcited by 1
  15. 2025HiGS: Hierarchical Generative Scene Framework for Multi-Step Associative Semantic Spatial Compositioncited by 1
  16. 2025Object Reconstruction under Occlusion with Generative Priors and Contact-induced Constraintscited by 1
  17. 2026RelaxFlow: Text-Driven Amodal 3D Generationcited by 1
  18. 2026Pano3DComposer: Feed-Forward Compositional 3D Scene Generation from Single Panoramic Imagecited by 1
  19. 2026SK-Adapter: Skeleton-Based Structural Control for Native 3D Generationcited by 1
  20. 2026AnyHand: A Large-Scale Synthetic Dataset for RGB(-D) Hand Pose Estimationcited by 1
  21. 2026Perceive-then-Plan: Layout-as-Policy for Monocular 3D Scene Layout Estimationcited by 1
  22. 2026CA-World: Multi-Object Counterfactual Alignment for Efficient Interactive-Ready Reconstructioncited by 1
  23. 2026FlowObject: Flow Steering for Bridging Generative Priors and Reconstruction Fidelitycited by 1
  24. 2026EditVerse3D: High-Quality 3D Object Editing with Region-Aware Learningcited by 1
  25. 2026GIF: Agentic Generation of Interactive and Functional Object Compositions for Robot Learningcited by 1
  26. 2025SCPainter: A Unified Framework for Realistic 3D Asset Insertion and Novel View Synthesiscited by 0
  27. 2026Object Pose Transformer: Unifying Unseen Object Pose Estimationcited by 0
  28. 2026Physically Grounded 3D Generative Reconstruction under Hand Occlusion using Proprioception and Multi-Contact Touchcited by 0
  29. 2026FurnSet: Exploiting Repeats for 3D Scene Reconstructioncited by 0
  30. 2026MAPRPose: Mask-Aware Proposal and Amodal Refinement for Multi-Object 6D Pose Estimationcited by 0
  31. 2026PersistGS: Differentiable Physics for Object Permanence in 4D Gaussian Splattingcited by 0
  32. 20263DMorph: Single-Image-Guided Local 3D Shape Editing and Morphingcited by 0
  33. 2026Occlusion-Robust Multi-Object Decoupling for Physics-Based Robotic Interactioncited by 0
  34. 2026Restore3D: Breathing Life into Broken Objects with Shape and Texture Restorationcited by 0
  35. 2026HIVE-3D: Hierarchical Voxel Enhancement for High-Quality 3D Scene Generationcited by 0
  36. 2026Scene-SAM3D: Multi-View Scene Asset Generation Without Fine-Tuningcited by 0
  37. 2026ObjectSplat: Improving Mesh Fidelity and Interactivity for 3D Scenes via Object-Level Mesh Splattingcited by 0
  38. 2026ZeroBot: Learning from Scratch in Minutes with Generative Real2Simcited by 0

50 most cited of 52 All 52 (JSON)