mlsanityData to 14 Oct 2026

ScanComplete: Large-Scale Scene Completion and Semantic Segmentation for 3D Scans

arXiv 1712.10215 ↗First posted 29 Dec 2017cs.CVcited by 103

Not on the map. Why?

Abstract (arXiv)

We introduce ScanComplete, a novel data-driven approach for taking an incomplete 3D scan of a scene as input and predicting a complete 3D model along with per-voxel semantic labels. The key contribution of our method is its ability to handle large scenes with varying spatial extent, managing the cubic growth in data size as scene size increases. To this end, we devise a fully-convolutional generative 3D CNN model whose filter kernels are invariant to the overall scene size. The model can be trained on scene subvolumes but deployed on arbitrarily large scenes at test time. In addition, we propose a coarse-to-fine inference strategy in order to produce high-resolution output while also leveraging large input context sizes. In an extensive series of experiments, we carefully evaluate different model design choices, considering both deterministic and probabilistic models for completion and semantic inference. Our results show that we outperform other methods not only in the size of the environments handled and processing efficiency, but also with regard to completion quality and semantic segmentation performance by a significant margin.

Key facts. Cited by 103 papers in the map (data through 14 Oct 2026). First posted on arXiv 29 Dec 2017.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 27; 2022: 22; 2023: 14; 2024: 17; 2025: 14; 2026: 9.

103 papers.

Sort:

By area:

  1. 2022OpenScene: 3D Scene Understanding with Open Vocabulariescited by 559
  2. 2021Deep Marching Tetrahedra: a Hybrid Representation for High-Resolution 3D Shape Synthesiscited by 460
  3. 2021Dense Depth Priors for Neural Radiance Fields from Sparse Input Viewscited by 302
  4. 2021Shape As Points: A Differentiable Poisson Solvercited by 187
  5. 20214DComplete: Non-Rigid Motion Estimation Beyond the Observable Surfacecited by 124
  6. 2021NeSF: Neural Semantic Fields for Generalizable Semantic Segmentation of 3D Scenescited by 119
  7. 2022Semantic Abstraction: Open-World 3D Scene Understanding from 2D Vision-Language Modelscited by 95
  8. 2022POCO: Point Convolution for Surface Reconstructioncited by 92
  9. 2023SCPNet: Semantic Scene Completion on Point Cloudcited by 80
  10. 20213D Semantic Scene Completion: a Surveycited by 74
  11. 2021Unsupervised 3D Shape Completion through GAN Inversioncited by 70
  12. 2021Panoptic 3D Scene Reconstruction From a Single RGB Imagecited by 70
38 more
  1. 2024Advances in 3D Generation: A Surveycited by 63
  2. 2021StyleMesh: Style Transfer for Indoor 3D Scene Reconstructionscited by 58
  3. 2021Semantics for Robotic Mapping, Perception and Interaction: A Surveycited by 56
  4. 2022Contrastive Boundary Learning for Point Cloud Segmentationcited by 56
  5. 2024InFusion: Inpainting 3D Gaussians via Learning Depth Completion from Diffusion Priorcited by 56
  6. 2024EmbodiedOcc: Embodied 3D Occupancy Prediction for Vision-based Online Scene Understandingcited by 45
  7. 2023PaSCo: Urban 3D Panoptic Scene Completion with Uncertainty Awarenesscited by 42
  8. 2024GScream: Learning 3D Geometry and Feature Consistent Gaussian Splatting for Object Removalcited by 39
  9. 2021VoRTX: Volumetric 3D Reconstruction With Transformers for Voxelwise View Selection and Fusioncited by 38
  10. 2022Deep Learning for Omnidirectional Vision: A Survey and New Perspectivescited by 38
  11. 2021Lite-HDSeg: LiDAR Semantic Segmentation Using Lite Harmonic Dense Convolutionscited by 37
  12. 2021Semantic Segmentation-assisted Scene Completion for LiDAR Point Cloudscited by 37
  13. 2021Deep Learning Based 3D Segmentation: A Surveycited by 30
  14. 2023SSC-RS: Elevate LiDAR Semantic Scene Completion with Representation Separation and BEV Fusioncited by 30
  15. 2024DepthLab: From Partial to Completecited by 27
  16. 2024S2O: Static to Openable Enhancement for Articulated 3D Objectscited by 22
  17. 2023SGRec3D: Self-Supervised 3D Scene Graph Learning via Object-Level Scene Reconstructioncited by 21
  18. 2022Point Scene Understanding via Disentangled Instance Mesh Reconstructioncited by 20
  19. 2023BUOL: A Bottom-Up Framework with Occupancy-aware Lifting for Panoptic 3D Scene Reconstruction From A Single Imagecited by 20
  20. 2023Marching-Primitives: Shape Abstraction from Signed Distance Functioncited by 19
  21. 2026Multiplayer Interactive World Models with Representation Autoencoderscited by 19
  22. 2022Learning a Structured Latent Space for Unsupervised Point Cloud Completioncited by 17
  23. 2024Open Vocabulary 3D Scene Understanding via Geometry Guided Self-Distillationcited by 17
  24. 2022Learning Local Displacements for Point Cloud Completioncited by 16
  25. 2024Dense Multimodal Alignment for Open-Vocabulary 3D Scene Understandingcited by 16
  26. 2022Multi-sensor large-scale dataset for multi-view 3D reconstructioncited by 14
  27. 2023ImGeoNet: Image-induced Geometry-aware Voxel Representation for Multi-view 3D Object Detectioncited by 14
  28. 2024AirPlanes: Accurate Plane Estimation via 3D-Consistent Embeddingscited by 13
  29. 2025Cross-Modal and Uncertainty-Aware Agglomeration for Open-Vocabulary 3D Scene Understandingcited by 13
  30. 2025ZeroGrasp: Zero-Shot Shape Reconstruction Enabled Robotic Graspingcited by 13
  31. 20213D Instance Segmentation of MVS Buildingscited by 12
  32. 2022Neural Scene Representation for Locomotion on Structured Terraincited by 12
  33. 2023RayDF: Neural Ray-surface Distance Fields with Multi-view Consistencycited by 12
  34. 2024Distilling Diffusion Models to Efficient 3D LiDAR Scene Completioncited by 10
  35. 2021Mask-ToF: Learning Microlens Masks for Flying Pixel Correction in Time-of-Flight Imagingcited by 9
  36. 2023Clutter Detection and Removal in 3D Scenes with View-Consistent Inpaintingcited by 9
  37. 2023GNeSF: Generalizable Neural Semantic Fieldscited by 9
  38. 2024Zero-Shot Multi-Object Scene Completioncited by 9

50 most cited of 103 All 103 (JSON)