mlsanityData to 14 Oct 2026

Attention-based Part Assembly for 3D Volumetric Shape Modeling

arXiv 2304.10986 ↗First posted 17 Apr 2023cs.CVcited by 3

Not on the map. Why?

Abstract (arXiv)

Modeling a 3D volumetric shape as an assembly of decomposed shape parts is much more challenging, but semantically more valuable than direct reconstruction from a full shape representation. The neural network needs to implicitly learn part relations coherently, which is typically performed by dedicated network layers that can generate transformation matrices for each part. In this paper, we propose a VoxAttention network architecture for attention-based part assembly. We further propose a variant of using channel-wise part attention and show the advantages of this approach. Experimental results show that our method outperforms most state-of-the-art methods for the part relation-aware 3D shape modeling task.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 37 of its 53 references that are in the map. First posted on arXiv 17 Apr 2023.

Builds on

37 of 53 references.

References · 37

  1. 2017Attention Is All You Needcited by 60,361
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2020End-to-End Object Detection with Transformerscited by 7,414
  4. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,897
  5. 2019DeepSDF: Learning Continuous Signed Distance Functions for Shape Representationcited by 2,689
  6. 2018Occupancy Networks: Learning 3D Reconstruction in Function Spacecited by 2,041
  7. 2020Point Transformercited by 1,470
  8. 2018Learning Implicit Fields for Generative Shape Modelingcited by 1,060
  9. 2020Pre-Trained Image Processing Transformercited by 893
  10. 2020PCT: Point cloud transformercited by 732
  11. 2018Pixel2Mesh: Generating 3D Mesh Models from Single RGB Imagescited by 651
  12. 20163D-R2N2: A Unified Approach for Single and Multi-view 3D Object Reconstructioncited by 588
25 more
  1. 2016Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modelingcited by 558
  2. 2019PointFlow: 3D Point Cloud Generation with Continuous Normalizing Flowscited by 488
  3. 2016OctNet: Learning Deep 3D Representations at High Resolutionscited by 342
  4. 2021Voxel Transformer for 3D Object Detectioncited by 247
  5. 2016Learning a Predictable and Generative Vector Representation for Objectscited by 173
  6. 2016Generative and Discriminative Voxel Modeling with Convolutional Neural Networkscited by 128
  7. 2019PQ-NET: A Generative Part Seq2Seq Network for 3D Shapescited by 119
  8. 2019Pixel2Mesh++: Multi-View 3D Mesh Generation via Deformationcited by 103
  9. 2019BAE-NET: Branched Autoencoder for Shape Co-Segmentationcited by 89
  10. 20173D-PRNN: Generating Shape Primitives with Recurrent Neural Networkscited by 77
  11. 2020Leveraging 2D Data to Learn Textured 3D Mesh Generationcited by 69
  12. 2019StructureNet: Hierarchical Graph Networks for 3D Shape Generationcited by 59
  13. 2018Multi-view Consistency as Supervisory Signal for Learning Shape and Pose Predictioncited by 52
  14. 2019C-Flow: Conditional Generative Flow Models for Images and 3D Point Cloudscited by 50
  15. 2020D2IM-Net: Learning Detail Disentangled Implicit Fields from Single Imagescited by 42
  16. 2019Learning Part Generation and Assembly for Structure-aware Shape Synthesiscited by 36
  17. 2019Composite Shape Modeling via Latent Space Factorizationcited by 31
  18. 2020COALESCE: Component Assembly by Learning to Synthesize Connectionscited by 30
  19. 2020SVGA-Net: Sparse Voxel-Graph Attention Network for 3D Object Detection from Point Cloudscited by 27
  20. 2018SCORES: Shape Composition with Recursive Substructure Priorscited by 24
  21. 2019StructEdit: Learning Structural Shape Variationscited by 23
  22. 2021EditVAE: Unsupervised Part-Aware Controllable 3D Point Cloud Shape Generationcited by 21
  23. 2018SAGNet:Structure-aware Generative Network for 3D-Shape Modelingcited by 20
  24. 2018Structure-Aware Shape Synthesiscited by 5
  25. 2023Self-Supervised Generative-Contrastive Learning of Multi-Modal Euclidean Input for 3D Shape Latent Representations: A Dynamic Switching Approachcited by 2

Built on by

Cited by year

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

3 papers.

  1. 2023Point Cloud Part Editing: Segmentation, Generation, Assembly, and Selectioncited by 4
  2. 2025Aligning Instance-Semantic Sparse Representation towards Unsupervised Object Segmentation and Shape Abstraction with Repeatable Primitivescited by 2
  3. 2024Exploiting Topological Priors for Boosting Point Cloud Generationcited by 0