mlsanityData to 14 Oct 2026

Jigsaw: Learning to Assemble Multiple Fractured Objects

arXiv 2305.17975 ↗First posted 29 May 2023cs.CVcited by 29

Not on the map. Why?

Abstract (arXiv)

Automated assembly of 3D fractures is essential in orthopedics, archaeology, and our daily life. This paper presents Jigsaw, a novel framework for assembling physically broken 3D objects from multiple pieces. Our approach leverages hierarchical features of global and local geometry to match and align the fracture surfaces. Our framework consists of four components: (1) front-end point feature extractor with attention layers, (2) surface segmentation to separate fracture and original parts, (3) multi-parts matching to find correspondences among fracture surface points, and (4) robust global alignment to recover the global poses of the pieces. We show how to jointly learn segmentation and matching and seamlessly integrate feature matching and rigidity constraints. We evaluate Jigsaw on the Breaking Bad dataset and achieve superior performance compared to state-of-the-art methods. Our method also generalizes well to diverse fracture modes, objects, and unseen instances. To the best of our knowledge, this is the first learning-based method designed specifically for 3D fracture assembly over multiple pieces. Our code is available at https://jiaxin-lu.github.io/Jigsaw/.

Key facts. Cited by 29 papers in the map (data through 14 Oct 2026). Builds on 25 of its 41 references that are in the map. First posted on arXiv 29 May 2023.

Builds on

25 of 41 references.

References · 25

  1. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,248
  2. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,401
  3. 2013Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distancescited by 2,316
  4. 2020Point Transformercited by 1,470
  5. 2019SuperGlue: Learning Feature Matching with Graph Neural Networkscited by 1,443
  6. 2021LoFTR: Detector-Free Local Feature Matching with Transformerscited by 984
  7. 2018Open3D: A Modern Library for 3D Data Processingcited by 721
  8. 2018PartNet: A Large-scale Benchmark for Fine-grained and Hierarchical Part-level 3D Object Understandingcited by 526
  9. 2020PREDATOR: Registration of 3D Point Clouds with Low Overlapcited by 290
  10. 2022Geometric Transformer for Fast and Robust Point Cloud Registrationcited by 202
  11. 2019PQ-NET: A Generative Part Seq2Seq Network for 3D Shapescited by 119
  12. 2019Learning Combinatorial Embedding Networks for Deep Graph Matchingcited by 99
13 more
  1. 2019Neural Graph Matching Network: Learning Lawler's Quadratic Assignment Problem with Extension to Hypergraph and Multiple-graph Matchingcited by 84
  2. 2020Deep Graph Matching via Blackbox Differentiation of Combinatorial Solverscited by 81
  3. 2021JoinABLe: Learning Bottom-up Assembly of Parametric CAD Jointscited by 78
  4. 2020Generative 3D Part Assembly via Dynamic Graph Learningcited by 67
  5. 2020Learning 3D Part Assembly from a Single Imagecited by 52
  6. 2020Shonan Rotation Averaging: Global Optimality by Surfing SO(p)ncited by 46
  7. 2022Neural Shape Mating: Self-Supervised Object Assembly with Adversarial Shape Priorscited by 43
  8. 2022Breaking Bad: A Dataset for Geometric Fracture and Reassemblycited by 41
  9. 2021RGL-NET: A Recurrent Graph Learning framework for Progressive Part Assemblycited by 36
  10. 2019Composite Shape Modeling via Latent Space Factorizationcited by 31
  11. 2020COALESCE: Component Assembly by Learning to Synthesize Connectionscited by 30
  12. 2021Point Cloud Registration using Representative Overlapping Pointscited by 5
  13. 2022Learning Universe Model for Partial Matching Networks over Multiple Graphscited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 8; 2025: 10; 2026: 11.

29 papers.

Sort:

By area:

  1. 2024Re-assembling the past: The RePAIR dataset and benchmark for real world 2D and 3D puzzle solvingcited by 19
  2. 2025GaussianArt: Unified Modeling of Geometry and Motion for Articulated Objectscited by 17
  3. 2024PuzzleFusion++: Auto-agglomerative 3D Fracture Assembly by Denoise and Verifycited by 15
  4. 2025Rectified Point Flow: Generic Point Cloud Pose Estimationcited by 14
  5. 2025GARF: Learning Generalizable 3D Reassembly for Real-World Fracturescited by 13
  6. 2025Two by Two: Learning Multi-Task Pairwise Objects Assembly for Generalizable Robot Manipulationcited by 13
  7. 2024SPAFormer: Sequential 3D Part Assembly with Transformerscited by 12
  8. 2025Assembler: Scalable 3D Part Assembly via Anchor Point Diffusioncited by 12
  9. 20243D Geometric Shape Assembly via Efficient Point Cloud Matchingcited by 9
  10. 2024A Survey on Computational Solutions for Reconstructing Complete Objects by Reassembling Their Fractured Partscited by 9
  11. 2024Jigsaw++: Imagining Complete Shape Priors for Object Reassemblycited by 6
  12. 2024Effective Protein-Protein Interaction Exploration with PPIretrievalcited by 5
17 more
  1. 2025BiAssemble: Learning Collaborative Affordance for Bimanual Geometric Assemblycited by 5
  2. 2025GarmageNet: A Multimodal Generative Framework for Sewing Pattern Design and Generic Garment Modelingcited by 4
  3. 2026TORA: Topological Representation Alignment for 3D Shape Assemblycited by 4
  4. 2024Geometric Point Attention Transformer for 3D Shape Reassemblycited by 3
  5. 2025Combinative Matching for Geometric Shape Assemblycited by 3
  6. 2026Interact3D: Compositional 3D Generation of Interactive Objectscited by 3
  7. 2025Equivariant Flow Matching for Point Cloud Assemblycited by 2
  8. 2025E-M3RF: An Equivariant Multimodal 3D Re-assembly Frameworkcited by 1
  9. 2026Learning Sewing Patterns via Latent Flow Matching of Implicit Fieldscited by 1
  10. 2026CRAG: Can 3D Generative Models Help 3D Assembly?cited by 1
  11. 2026AutoSew: A Geometric Approach to Stitching Prediction with Graph Neural Networkscited by 0
  12. 2026Transparent Fragments Contour Estimation via Visual-Tactile Fusion for Autonomous Reassemblycited by 0
  13. 2026LAMP: Lift Image-Editing as General 3D Priors for Open-world Manipulationcited by 0
  14. 2026AssemblyWorld: Rethinking 3D Assembly with General-Purpose Agentscited by 0
  15. 2026Automatic assembly modeling through attention-based assembly feature mate prediction and assembly feature pattern recognitioncited by 0
  16. 2026LLM2R: From Large Language Models to 3D Reassemblerscited by 0
  17. 2026PhySGA: A Physics-Space Grounded Agentic Framework for Generalizable and Interactive 3D Compositioncited by 0