mlsanityData to 7 Oct 2026

Scan2Part: Fine-grained and Hierarchical Part-level Understanding of Real-World 3D Scans

arXiv 2206.02366 ↗First posted 6 Jun 2022cs.CVcited by 3

Not on the map. Why?

Abstract (arXiv)

We propose Scan2Part, a method to segment individual parts of objects in real-world, noisy indoor RGB-D scans. To this end, we vary the part hierarchies of objects in indoor scenes and explore their effect on scene understanding models. Specifically, we use a sparse U-Net-based architecture that captures the fine-scale detail of the underlying 3D scan geometry by leveraging a multi-scale feature hierarchy. In order to train our method, we introduce the Scan2Part dataset, which is the first large-scale collection providing detailed semantic labels at the part level in the real-world setting. In total, we provide 242,081 correspondences between 53,618 PartNet parts of 2,477 ShapeNet objects and 1,506 ScanNet scenes, at two spatial resolutions of 2 cm3 and 5 cm3. As output, we are able to predict fine-grained per-object part labels, even when the geometry is coarse or partially missing.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 45 of its 66 references that are in the map. First posted on arXiv 6 Jun 2022.

Builds on

45 of 66 references.

References · 45

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  3. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  4. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,088
  5. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,880
  6. 2016Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotationscited by 2,646
  7. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,220
  8. 2017Matterport3D: Learning from RGB-D Data in Indoor Environmentscited by 1,354
  9. 20194D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networkscited by 1,201
  10. 2019The Replica Dataset: A Digital Replica of Indoor Spacescited by 885
  11. 20173D Semantic Segmentation with Submanifold Sparse Convolutional Networkscited by 746
  12. 2019Revisiting Point Cloud Classification: A New Benchmark Dataset and Classification Model on Real-World Datacited by 532
33 more
  1. 2018PartNet: A Large-scale Benchmark for Fine-grained and Hierarchical Part-level 3D Object Understandingcited by 521
  2. 2016Semantic Scene Completion from a Single Depth Imagecited by 468
  3. 2017Scene Graph Generation by Iterative Message Passingcited by 462
  4. 2017Joint 2D-3D-Semantic Data for Indoor Scene Understandingcited by 404
  5. 2018Image Generation from Scene Graphscited by 324
  6. 20183D-SIS: 3D Semantic Instance Segmentation of RGB-D Scanscited by 216
  7. 2017SGPN: Similarity Group Proposal Network for 3D Point Cloud Instance Segmentationcited by 168
  8. 2019Learning Object Bounding Boxes for 3D Instance Segmentation on Point Cloudscited by 156
  9. 2018GSPN: Generative Shape Proposal Network for 3D Instance Segmentation in Point Cloudcited by 145
  10. 2017Escape from Cells: Deep Kd-Networks for the Recognition of 3D Point Cloud Modelscited by 142
  11. 2018Scan2CAD: Learning CAD Model Alignment in RGB-D Scanscited by 142
  12. 20203D-MPA: Multi Proposal Aggregation for 3D Semantic Instance Segmentationcited by 140
  13. 2017Semantic Instance Segmentation with a Discriminative Loss Functioncited by 118
  14. 20183DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentationcited by 118
  15. 2019PQ-NET: A Generative Part Seq2Seq Network for 3D Shapescited by 118
  16. 2019BAE-NET: Branched Autoencoder for Shape Co-Segmentationcited by 89
  17. 2018InteriorNet: Mega-scale Multi-sensor Photo-realistic Indoor Scenes Datasetcited by 87
  18. 2019Associatively Segmenting Instances and Semantics in Point Cloudscited by 86
  19. 2017Intel RealSense Stereoscopic Depth Camerascited by 82
  20. 20193D Instance Segmentation via Multi-Task Metric Learningcited by 79
  21. 2017Superhuman Accuracy on the SNEMI3D Connectomics Challengecited by 64
  22. 2019JSIS3D: Joint Semantic-Instance Segmentation of 3D Point Clouds with Multi-Task Pointwise Networks and Multi-Value Conditional Random Fieldscited by 62
  23. 2019Point Cloud Instance Segmentation using Probabilistic Embeddingscited by 52
  24. 2015SceneNet: Understanding Real World Indoor Scenes With Synthetic Datacited by 29
  25. 2019MASC: Multi-scale Affinity with Sparse Convolution for 3D Instance Segmentationcited by 25
  26. 2019'Zhores' -- Petaflops supercomputer for data-driven modeling, machine learning and artificial intelligence installed in Skolkovo Institute of Science and Technologycited by 23
  27. 2020PT2PC: Learning to Generate 3D Point Cloud Shapes from Part Tree Conditionscited by 20
  28. 2018Real-time Progressive 3D Semantic Segmentation for Indoor Scenecited by 18
  29. 2019The RobotriX: An eXtremely Photorealistic and Very-Large-Scale Indoor Dataset of Sequences with Robot Trajectories and Interactionscited by 14
  30. 2017Learning Hierarchical Shape Segmentation and Labeling from Online Repositoriescited by 11
  31. 2020Towards Part-Based Understanding of RGB-D Scanscited by 9
  32. 20193D Graph Embedding Learning with a Structure-aware Loss Function for Point Cloud Semantic Instance Segmentationcited by 4
  33. 2019AdaCoSeg: Adaptive Shape Co-Segmentation with Group Consistency Losscited by 2

Built on by

Cited by year

Citing papers per year: 2022: 1; 2023: 1; 2024: 0; 2025: 1.

3 papers.

  1. 2022PartSLIP: Low-Shot Part Segmentation for 3D Point Clouds via Pretrained Image-Language Modelscited by 117
  2. 2023PartSLIP++: Enhancing Low-Shot 3D Part Segmentation via Multi-View Instance Segmentation and Maximum Likelihood Estimationcited by 37
  3. 2025PinPoint3D: Fine-Grained 3D Part Segmentation from a Few Clickscited by 0