Skip to main content

DeepMend: Learning Occupancy Functions to Represent Shape for Repair

  • Conference paper
  • First Online:
Computer Vision – ECCV 2022 (ECCV 2022)

Abstract

We present DeepMend, a novel approach to reconstruct resto- rations to fractured shapes using learned occupancy functions. Existing shape repair approaches predict low-resolution voxelized restorations or smooth restorations, or require symmetries or access to a pre-existing complete oracle. We represent the occupancy of a fractured shape as the conjunction of the occupancy of an underlying complete shape and a break surface, which we model as functions of latent codes using neural networks. Given occupancy samples from a fractured shape, we estimate latent codes using an inference loss augmented with novel penalties to avoid empty or voluminous restorations. We use the estimated codes to reconstruct a restoration shape. We show results with simulated fractures on synthetic and real-world scanned objects, and with scanned real fractured mugs. Compared to existing approaches and to two baseline methods, our work shows state-of-the-art results in accuracy and avoiding restoration artifacts over non-fracture regions of the fractured shape.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD 89.00
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 119.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Notes

  1. 1.

    Hereafter, we drop ‘set’ from references to C, F, and R, and refer to them as shapes.

References

  1. Achlioptas, P., Diamanti, O., Mitliagkas, I., Guibas, L.: Learning representations and generative models for 3d point clouds. In: International Conference On Machine Learning. PMLR (2018)

    Google Scholar 

  2. Attene, M.: A lightweight approach to repairing digitized polygon meshes. Visual Comput. 26(11), 1393–1406 (2010)

    Article  Google Scholar 

  3. Brock, A., Lim, T., Ritchie, J.M., Weston, N.: Generative and discriminative voxel modeling with convolutional neural networks. arXiv preprint arXiv:1608.04236 1(1), 1–9 (2016)

  4. Chabra, R., Lenssen, J.E., Ilg, E., Schmidt, T., Straub, J., Lovegrove, S., Newcombe, R.: Deep local shapes: learning local SDF priors for detailed 3D reconstruction. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12374, pp. 608–625. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58526-6_36

    Chapter  Google Scholar 

  5. Chang, A.X., et al.: ShapeNet: An Information-Rich 3D Model Repository. Technology Report. arXiv:1512.03012 [cs.GR], Stanford University – Princeton University – Toyota Technological Institute at Chicago (2015)

  6. Chen, Y.C., Li, H., Turpin, D., Jacobson, A., Garg, A.: Neural shape mating: self-supervised object assembly with adversarial shape priors. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 12724–12733 (2022)

    Google Scholar 

  7. Chen, Z., Zhang, H.: Learning implicit fields for generative shape modeling. In: Proceedings of CVPR, pp. 5939–5948. IEEE, Piscataway, NJ (2019)

    Google Scholar 

  8. Chibane, J., Alldieck, T., Pons-Moll, G.: Implicit functions in feature space for 3d shape reconstruction and completion. In: Proceedings of CVPR, pp. 6970–6981. IEEE, Piscataway, NJ (2020)

    Google Scholar 

  9. Dai, A., Diller, C., Nießner, M.: SG-NN: sparse generative neural networks for self-supervised scene completion of RGB-D scans. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 849–858 (2020)

    Google Scholar 

  10. Dai, A., Ritchie, D., Bokeloh, M., Reed, S., Sturm, J., Nießner, M.: ScanComplete: large-scale scene completion and semantic segmentation for 3d scans. In: Proceedings of CVPR, pp. 4578–4587. IEEE, Piscataway, NJ (2018)

    Google Scholar 

  11. Dai, A., Ruizhongtai Qi, C., Nießner, M.: Shape completion using 3d-encoder-predictor CNNs and shape synthesis. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 5868–5877 (2017)

    Google Scholar 

  12. Duan, Y., Zhu, H., Wang, H., Yi, L., Nevatia, R., Guibas, L.J.: Curriculum DeepSDF. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12353, pp. 51–67. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58598-3_4

    Chapter  Google Scholar 

  13. Duchon, J.: Splines minimizing rotation-invariant semi-norms in Sobolev spaces. In: Schempp, W., Zeller, K. (eds) Constructive Theory of Functions of Several Variables. Lecture Notes in Mathematics, vol. 571. Springer, Berlin, Heidelberg (1977). https://doi.org/10.1007/BFb0086566

  14. Duggal, S., et al.: Mending neural implicit modeling for 3d vehicle reconstruction in the wild. In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pp. 1900–1909 (2022)

    Google Scholar 

  15. Genova, K., Cole, F., Sud, A., Sarna, A., Funkhouser, T.: Local deep implicit functions for 3d shape. In: Proceedings of CVPR, pp. 4857–4866. IEEE, Piscataway, NJ (2020)

    Google Scholar 

  16. Genova, K., Cole, F., Vlasic, D., Sarna, A., Freeman, W.T., Funkhouser, T.: Learning shape templates with structured implicit functions. In: Proceedings of CVPR, pp. 7154–7164. IEEE, Piscataway, NJ (2019)

    Google Scholar 

  17. Gregor, R., Sipiran, I., Papaioannou, G., Schreck, T., Andreadis, A., Mavridis, P.: Towards automated 3d reconstruction of defective cultural heritage objects. In: GCH, pp. 135–144. EUROGRAPHICS, Geneva, Switzerland (2014)

    Google Scholar 

  18. Groueix, T., Fisher, M., Kim, V.G., Russell, B.C., Aubry, M.: A papier-mâché approach to learning 3d surface generation. In: Proceedings of CVPR, pp. 216–224. IEEE, Piscataway, NJ (2018)

    Google Scholar 

  19. Gupta, M.M., Qi, J.: Theory of T-norms and fuzzy inference methods. Fuzzy Sets Syst. 40(3), 431–450 (1991)

    Article  MathSciNet  MATH  Google Scholar 

  20. Han, X., Li, Z., Huang, H., Kalogerakis, E., Yu, Y.: High-resolution shape completion using deep neural networks for global structure and local geometry inference. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 85–93 (2017)

    Google Scholar 

  21. Hao, Z., Averbuch-Elor, H., Snavely, N., Belongie, S.: DualSDF: semantic shape manipulation using a two-level representation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 7631–7641 (2020)

    Google Scholar 

  22. Hermoza, R., Sipiran, I.: 3d reconstruction of incomplete archaeological objects using a generative adversarial network. In: Proceedings of Computer Graphics International, pp. 5–11. ACM, New York, NY (2018)

    Google Scholar 

  23. Jia, M., Kyan, M.: Learning occupancy function from point clouds for surface reconstruction. arXiv preprint arXiv:2010.11378 (2020)

  24. Koutsoudis, A., Pavlidis, G., Arnaoutoglou, F., Tsiafakis, D., Chamzas, C.: QP: a tool for generating 3d models of ancient Greek pottery. J. Cult. Herit. 10(2), 281–295 (2009)

    Article  Google Scholar 

  25. Lamb, N., Banerjee, S., Banerjee, N.K.: MendNet: Restoration of Fractured Shapes Using Learned Occupancy Functions. Computer Graphics Forum (2022)

    Google Scholar 

  26. Lamb, N., Banerjee, S., Banerjee, N.K.: Automated reconstruction of smoothly joining 3d printed restorations to fix broken objects. In: Proceedings of SCF, pp. 1–12. ACM, New York, NY (2019)

    Google Scholar 

  27. Lamb, N., Wiederhold, N., Lamb, B., Banerjee, S., Banerjee, N.K.: Using learned visual and geometric features to retrieve complete 3d proxies for broken objects. In: Proceedings of SCF, pp. 1–15. ACM, New York, NY (2021)

    Google Scholar 

  28. Liao, Y., Donne, S., Geiger, A.: Deep marching cubes: learning explicit surface representations. In: Proceedings of CVPR, pp. 2916–2925. IEEE, Piscataway, NJ (2018)

    Google Scholar 

  29. Lin, C.H., Wang, C., Lucey, S.: SDF-SRN: learning signed distance 3d object reconstruction from static images. arXiv preprint arXiv:2010.10505 (2020)

  30. Lionar, S., Emtsev, D., Svilarkovic, D., Peng, S.: Dynamic plane convolutional occupancy networks. In: Proceedings of WACV, pp. 1829–1838. IEEE, Piscataway, NJ (2021)

    Google Scholar 

  31. Liu, B., Wang, M., Niu, X., Wang, S., Zhang, S., Zhang, J.: A fragment fracture surface segmentation method based on learning of local geometric features on margins used for automatic utensil reassembly. Comput. Aid. Des. 132, 102963 (2021)

    Article  MathSciNet  Google Scholar 

  32. Liu, M., Sheng, L., Yang, S., Shao, J., Hu, S.M.: Morphing and sampling network for dense point cloud completion. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 34, pp. 11596–11603. AAAI, Menlo Park, CA (2020)

    Google Scholar 

  33. Lorensen, W.E., Cline, H.E.: Marching cubes: a high resolution 3d surface construction algorithm. ACM SIGGRAPH Comput. Graph. 21(4), 163–169 (1987)

    Article  Google Scholar 

  34. Ma, B., Han, Z., Liu, Y.S., Zwicker, M.: Neural-pull: learning signed distance functions from point clouds by learning to pull space onto surfaces. arXiv preprint arXiv:2011.13495 (2020)

  35. Mescheder, L., Oechsle, M., Niemeyer, M., Nowozin, S., Geiger, A.: Occupancy networks: learning 3d reconstruction in function space. In: Proceedings of CVPR, pp. 4460–4470. IEEE, Piscataway, NJ (2019)

    Google Scholar 

  36. Pan, L., et al.: Variational relational point completion network. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 8524–8533 (2021)

    Google Scholar 

  37. Papaioannou, G., et al.: From reassembly to object completion: a complete systems pipeline. J. Comput. Cult. Herit. 10(2), 1–22 (2017)

    Article  Google Scholar 

  38. Park, J.J., Florence, P., Straub, J., Newcombe, R., Lovegrove, S.: DeepSDF: learning continuous signed distance functions for shape representation. In: Proceedings of CVPR, pp. 165–174. IEEE, Piscataway, NJ (2019)

    Google Scholar 

  39. Peng, S., Niemeyer, M., Mescheder, L., Pollefeys, M., Geiger, A.: Convolutional occupancy networks. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12348, pp. 523–540. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58580-8_31

    Chapter  Google Scholar 

  40. Poursaeed, O., Fisher, M., Aigerman, N., Kim, V.G.: Coupling explicit and implicit surface representations for generative 3D modeling. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12355, pp. 667–683. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58607-2_39

    Chapter  Google Scholar 

  41. Qi, C.R., Su, H., Mo, K., Guibas, L.J.: Pointnet: deep learning on point sets for 3d classification and segmentation. In: Proceedings of CVPR, pp. 652–660. IEEE, Piscataway, NJ (2017)

    Google Scholar 

  42. Google Research: Google scanned objects, August 2021. https://fuel.ignitionrobotics.org/1.0/Google%20Research/fuel/collections/Google%20Scanned%20Objects

  43. Sarmad, M., Lee, H.J., Kim, Y.M.: RL-GAN-Net: a reinforcement learning agent controlled GAN network for real-time point cloud shape completion. In: Proceedings of CVPR, pp. 5898–5907. IEEE, Piscataway, NJ (2019)

    Google Scholar 

  44. Sharma, A., Grau, O., Fritz, M.: VConv-DAE: deep volumetric shape learning without object labels. In: Hua, G., Jégou, H. (eds.) ECCV 2016. LNCS, vol. 9915, pp. 236–250. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-49409-8_20

    Chapter  Google Scholar 

  45. Sitzmann, V., Chan, E.R., Tucker, R., Snavely, N., Wetzstein, G.: MetaSDF: meta-learning signed distance functions. arXiv preprint arXiv:2006.09662 (2020)

  46. Smith, E.J., Meger, D.: Improved adversarial systems for 3d object generation and reconstruction. In: Conference on Robot Learning, pp. 87–96. PMLR, Cambridge, UK (2017)

    Google Scholar 

  47. Son, H., Kim, Y.M.: SAUM: Symmetry-aware upsampling module for consistent point cloud completion. In: Ishikawa, H., Liu, C.-L., Pajdla, T., Shi, J. (eds.) ACCV 2020. LNCS, vol. 12622, pp. 158–174. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-69525-5_10

    Chapter  Google Scholar 

  48. Stutz, D., Geiger, A.: Learning 3d shape completion under weak supervision. CoRR abs/1805.07290 (2018). https://arxiv.org/abs/1805.07290

  49. Sulzer, R., Landrieu, L., Boulch, A., Marlet, R., Vallet, B.: Deep surface reconstruction from point clouds with visibility information. arXiv preprint arXiv:2202.01810 (2022)

  50. Tang, J., Lei, J., Xu, D., Ma, F., Jia, K., Zhang, L.: Sign-agnostic CoNet: learning implicit surface reconstructions by sign-agnostic optimization of convolutional occupancy networks. arXiv preprint arXiv:2105.03582 (2021)

  51. Tretschk, E., Tewari, A., Golyanik, V., Zollhöfer, M., Stoll, C., Theobalt, C.: PatchNets: patch-based generalizable deep implicit 3d shape representations. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12361, pp. 293–309. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58517-4_18

    Chapter  Google Scholar 

  52. Wang, H., Zang, Y., Liang, F., Dong, Z., Fan, H., Yang, B.: A probabilistic method for fractured cultural relics automatic reassembly. J. Comput. Cult. Herit. (JOCCH) 14(1), 1–25 (2021)

    Article  Google Scholar 

  53. Wu, J., Zhang, C., Xue, T., Freeman, W.T., Tenenbaum, J.B.: Learning a probabilistic latent space of object shapes via 3d generative-adversarial modeling. In: Proceedings of NeurIPS, pp. 82–90. Neural Information Processing Systems, San Diego, CA (2016)

    Google Scholar 

  54. Xiao, Y., Xu, J., Gao, S.: Taylorimnet for fast 3d shape reconstruction based on implicit surface function. arXiv preprint arXiv:2201.06845 (2022)

  55. Xu, Yifan, Fan, Tianqi, Yuan, Yi., Singh, Gurprit: Ladybird: quasi-monte Carlo sampling for deep implicit field based 3d reconstruction with symmetry. In: Vedaldi, Andrea, Bischof, Horst, Brox, Thomas, Frahm, Jan-Michael. (eds.) ECCV 2020. LNCS, vol. 12346, pp. 248–263. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58452-8_15

    Chapter  Google Scholar 

  56. Yan, S., Yang, Z., Li, H., Guan, L., Kang, H., Hua, G., Huang, Q.: Implicit autoencoder for point cloud self-supervised representation learning. arXiv preprint arXiv:2201.00785 (2022)

  57. Yan, X., Lin, L., Mitra, N.J., Lischinski, D., Cohen-Or, D., Huang, H.: ShapeFormer: Transformer-based shape completion via sparse representation. arXiv preprint arXiv:2201.10326 (2022)

  58. Yang, M., Wen, Y., Chen, W., Chen, Y., Jia, K.: Deep optimized priors for 3d shape modeling and reconstruction. In: Proceedings of CVPR, pp. 3269–3278. IEEE, Piscataway, NJ (2021)

    Google Scholar 

  59. Yi, L., Gong, B., Funkhouser, T.: Complete & label: a domain adaptation approach to semantic segmentation of lidar point clouds. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 15363–15373 (2021)

    Google Scholar 

  60. Yu, Q., Yang, C., Wei, H.: Part-wise Atlasnet for 3d point cloud reconstruction from a single image. Knowledge-Based Systems, p. 108395 (2022)

    Google Scholar 

  61. Yuan, W., Khot, T., Held, D., Mertz, C., Hebert, M.: PCN: point completion network. In: 2018 International Conference on 3D Vision (3DV), pp. 728–737. IEEE (2018)

    Google Scholar 

  62. Zheng, Z., Yu, T., Dai, Q., Liu, Y.: Deep implicit templates for 3d shape representation. In: Proceedings of CVPR, pp. 1429–1439. IEEE, Piscataway, NJ (2021)

    Google Scholar 

Download references

Acknowledgments

This work is supported by the National Science Foundation (NSF) Graduate Research Fellowship Program under Grant 1945954. We thank the Clarkson University Office of Information Technology for the use of the ACRES GPU server sponsored by NSF Grant 1925596.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Nikolas Lamb .

Editor information

Editors and Affiliations

1 Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary material 1 (pdf 7440 KB)

Rights and permissions

Reprints and permissions

Copyright information

© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG

About this paper

Check for updates. Verify currency and authenticity via CrossMark

Cite this paper

Lamb, N., Banerjee, S., Banerjee, N.K. (2022). DeepMend: Learning Occupancy Functions to Represent Shape for Repair. In: Avidan, S., Brostow, G., Cissé, M., Farinella, G.M., Hassner, T. (eds) Computer Vision – ECCV 2022. ECCV 2022. Lecture Notes in Computer Science, vol 13663. Springer, Cham. https://doi.org/10.1007/978-3-031-20062-5_25

Download citation

  • DOI: https://doi.org/10.1007/978-3-031-20062-5_25

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-031-20061-8

  • Online ISBN: 978-3-031-20062-5

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics