mlsanityData to 14 Oct 2026

Relation3D: Enhancing Relation Modeling for Point Cloud Instance Segmentation

arXiv 2506.17891 ↗First posted 22 Jun 2025cs.CVcited by 11

Not on the map. Why?

Abstract (arXiv)

3D instance segmentation aims to predict a set of object instances in a scene, representing them as binary foreground masks with corresponding semantic labels. Currently, transformer-based methods are gaining increasing attention due to their elegant pipelines and superior predictions. However, these methods primarily focus on modeling the external relationships between scene features and query features through mask attention. They lack effective modeling of the internal relationships among scene features as well as between query features. In light of these disadvantages, we propose Relation3D: Enhancing Relation Modeling for Point Cloud Instance Segmentation. Specifically, we introduce an adaptive superpoint aggregation module and a contrastive learning-guided superpoint refinement module to better represent superpoint features (scene features) and leverage contrastive learning to guide the updates of these features. Furthermore, our relation-aware self-attention mechanism enhances the capabilities of modeling relationships between queries by incorporating positional and geometric relationships into the self-attention mechanism. Extensive experiments on the ScanNetV2, ScanNet++, ScanNet200 and S3DIS datasets demonstrate the superior performance of Relation3D.

Key facts. Cited by 11 papers in the map (data through 14 Oct 2026). Builds on 47 of its 62 references that are in the map. First posted on arXiv 22 Jun 2025.

Builds on

47 of 62 references.

References · 47

  1. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  2. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 12,049
  3. 2017Mask R-CNNcited by 7,277
  4. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,248
  5. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,401
  6. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,117
  7. 2020Supervised Contrastive Learningcited by 2,829
  8. 2021Masked-attention Mask Transformer for Universal Image Segmentationcited by 2,351
  9. 2020Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformerscited by 1,369
  10. 20234D Gaussian Splatting for Real-Time Dynamic Scene Renderingcited by 1,157
  11. 2021Per-Pixel Classification is Not All You Need for Semantic Segmentationcited by 1,152
  12. 2023ScanNet++: A High-Fidelity Dataset of 3D Indoor Scenescited by 764
35 more
  1. 2023Dynamic 3D Gaussians: Tracking by Persistent Dynamic View Synthesiscited by 737
  2. 2017Frustum PointNets for 3D Object Detection from RGB-D Datacited by 640
  3. 2019A Survey of Autonomous Driving: Common Practices and Emerging Technologiescited by 379
  4. 2022Mask DINO: Towards A Unified Transformer-based Framework for Object Detection and Segmentationcited by 344
  5. 2020PointGroup: Dual-Set Point Grouping for 3D Instance Segmentationcited by 321
  6. 2022Language-Grounded Indoor 3D Semantic Segmentation in the Wildcited by 246
  7. 20183D-SIS: 3D Semantic Instance Segmentation of RGB-D Scanscited by 217
  8. 2022SoftGroup for 3D Instance Segmentation on Point Cloudscited by 200
  9. 2020OccuSeg: Occupancy-aware 3D Instance Segmentationcited by 177
  10. 2019Learning Object Bounding Boxes for 3D Instance Segmentation on Point Cloudscited by 156
  11. 2018GSPN: Generative Shape Proposal Network for 3D Instance Segmentation in Point Cloudcited by 146
  12. 20203D-MPA: Multi Proposal Aggregation for 3D Semantic Instance Segmentationcited by 141
  13. 2018Towards End-to-End Lane Detection: an Instance Segmentation Approachcited by 129
  14. 2021Hierarchical Aggregation for 3D Instance Segmentationcited by 127
  15. 2022Superpoint Transformer for 3D Scene Instance Segmentationcited by 116
  16. 2021Instance Segmentation in 3D Scenes using Semantic Superpoint Tree Networkscited by 99
  17. 20193D Instance Segmentation via Multi-Task Metric Learningcited by 80
  18. 2024Align3R: Aligned Monocular Depth Estimation for Dynamic Videoscited by 79
  19. 2020DyCo3D: Robust Instance Segmentation of 3D Point Clouds through Dynamic Convolutioncited by 75
  20. 2024MotionGS: Exploring Explicit Motion Guidance for Deformable 3D Gaussian Splattingcited by 63
  21. 2023ISBNet: a 3D Point Cloud Instance Segmentation Network with Instance-aware Sampling and Box-aware Dynamic Convolutioncited by 53
  22. 2019Point Cloud Instance Segmentation using Probabilistic Embeddingscited by 52
  23. 2023Mask-Attention-Free Transformer for 3D Instance Segmentationcited by 46
  24. 2024EmbodiedSAM: Online Segment Any 3D Thing in Real Timecited by 40
  25. 20223D Instances as 1D Kernelscited by 36
  26. 2022MaskGroup: Hierarchical Point Grouping and Masking for 3D Instance Segmentationcited by 30
  27. 2024Relation DETR: Exploring Explicit Position Relation Prior for Object Detectioncited by 28
  28. 2024DN-4DGS: Denoised Deformable Network with Temporal-Spatial Aggregation for Dynamic Scene Renderingcited by 28
  29. 2023Spherical Mask: Coarse-to-Fine 3D Point Cloud Instance Segmentation with Spherical Representationcited by 24
  30. 2019Rescan: Inductive Instance Segmentation for Indoor RGBD Scanscited by 12
  31. 2024BSNet: Box-Supervised Simulation-assisted Mean Teacher for 3D Instance Segmentationcited by 9
  32. 2024Diff3DETR:Agent-based Diffusion Model for Semi-supervised 3D Object Detectioncited by 7
  33. 2025SAS: Segment Any 3D Scene with Integrated 2D Priorscited by 6
  34. 2021CORE-Text: Improving Scene Text Detection with Contrastive Relational Reasoningcited by 4
  35. 2025Beyond the Final Layer: Hierarchical Query Fusion Transformer with Agent-Interpolation Initialization for 3D Instance Segmentationcited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 10.

11 papers.

  1. 2026Multiplayer Interactive World Models with Representation Autoencoderscited by 19
  2. 2025SegDINO3D: 3D Instance Segmentation Empowered by Both Image-Level and Object-Level 2D Featurescited by 4
  3. 2026Volume Transformer: Revisiting Vanilla Transformers for 3D Scene Understandingcited by 4
  4. 2026LaSSM: Efficient Semantic-Spatial Query Decoding via Local Aggregation and State Space Models for 3D Instance Segmentationcited by 3
  5. 2026SegVGGT: Joint 3D Reconstruction and Instance Segmentation from Multi-View Imagescited by 3
  6. 2026SegmentAnyTreeV2: Scaling Transformer-Based Tree Instance Segmentation Across Sensors, Platforms, and Forestscited by 3
  7. 2026ReScene4D: Temporally Consistent Semantic Instance Segmentation of Evolving Indoor 3D Scenescited by 2
  8. 2026GeoGuide: Hierarchical Geometric Guidance for Open-Vocabulary 3D Semantic Segmentationcited by 1
  9. 2026AQ3D: Adaptive Query Transformer for 3D Instance Segmentationcited by 0
  10. 2026Instance Segmentation and Fine-grained Classification for Urban Buildings with Adaptive Region Dividing and Spatially-Supervised Contrastive Learningcited by 0
  11. 2026Geometry-Guided Semantic Reconstruction for 3D Instance Segmentationcited by 0