Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
166 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
42 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

RDA-INR: Riemannian Diffeomorphic Autoencoding via Implicit Neural Representations (2305.12854v3)

Published 22 May 2023 in eess.IV, cs.CV, and cs.GR

Abstract: Diffeomorphic registration frameworks such as Large Deformation Diffeomorphic Metric Mapping (LDDMM) are used in computer graphics and the medical domain for atlas building, statistical latent modeling, and pairwise and groupwise registration. In recent years, researchers have developed neural network-based approaches regarding diffeomorphic registration to improve the accuracy and computational efficiency of traditional methods. In this work, we focus on a limitation of neural network-based atlas building and statistical latent modeling methods, namely that they either are (i) resolution dependent or (ii) disregard any data- or problem-specific geometry needed for proper mean-variance analysis. In particular, we overcome this limitation by designing a novel encoder based on resolution-independent implicit neural representations. The encoder achieves resolution invariance for LDDMM-based statistical latent modeling. Additionally, the encoder adds LDDMM Riemannian geometry to resolution-independent deep learning models for statistical latent modeling. We investigate how the Riemannian geometry improves latent modeling and is required for a proper mean-variance analysis. To highlight the benefit of resolution independence for LDDMM-based data variability modeling, we show that our approach outperforms current neural network-based LDDMM latent code models. Our work paves the way for more research into how Riemannian geometry, shape respectively image analysis, and deep learning can be combined.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (88)
  1. Resnet-lddmm: Advancing the lddmm framework using deep residual networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 45(3):3707–3720, 2022.
  2. Ankita Joshi and Yi Hong. Diffeomorphic image registration using lipschitz continuous residual networks. In Proceedings of The 5th International Conference on Medical Imaging with Deep Learning, volume 172 of Proceedings of Machine Learning Research, pages 605–617. PMLR, 2022.
  3. Learning a probabilistic model for diffeomorphic registration. IEEE transactions on medical imaging, 38(9):2165–2176, 2019.
  4. Networks for joint affine and non-parametric image registration. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 4224–4233, 2019.
  5. Quicksilver: Fast predictive image registration–a deep learning approach. NeuroImage, 158:378–396, 2017.
  6. Metric learning for image registration. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 8463–8472, 2019.
  7. Learning conditional deformable templates with convolutional networks. In Advances in Neural Information Processing Systems, volume 32. Curran Associates, Inc., 2019.
  8. Diffeomorphic image registration with neural velocity field. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), pages 1869–1879, 2023.
  9. Nephi: Neural deformation fields for approximately diffeomorphic medical image registration. arXiv preprint arXiv:2309.07322, 2023.
  10. Homeomorphic image registration via conformal-invariant hyperelastic regularisation. arXiv preprint arXiv:2303.08113, 2023.
  11. Fourier-net+: Leveraging band-limited representation for efficient 3d medical image registration. arXiv preprint arXiv:2307.02997, 2023.
  12. Deepflash: An efficient network for learning-based medical image registration. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 4444–4452, 2020.
  13. Tony C. W. Mok and Albert C. S. Chung. Large deformation diffeomorphic image registration with laplacian pyramid networks. In Medical Image Computing and Computer Assisted Intervention (MICCAI), pages 211–221. Springer, 2020.
  14. Aladdin: Joint atlas building and diffeomorphic registration learning with pairwise alignment. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 20784–20793, 2022.
  15. Diffeomorphic autoencoders for lddmm atlas building. 2018.
  16. Learning low-dimensional representations of shape data sets with diffeomorphic autoencoders. In Information Processing in Medical Imaging, pages 195–207. Springer, 2019.
  17. Geo-sic: Learning deformable geometric shapes in deep image classifiers. In Advances in Neural Information Processing Systems, volume 35, pages 27994–28007. Curran Associates, Inc., 2022.
  18. Neurepdiff: Neural operators to predict geodesics in deformation spaces. In Information Processing in Medical Imaging, pages 588–600. Springer, 2023.
  19. Landmark-free statistical shape modeling via neural flow deformations. In Medical Image Computing and Computer Assisted Intervention (MICCAI), pages 453–463. Springer, 2022.
  20. Neural mesh flow: 3d manifold mesh generation via diffeomorphic flows. In Advances in Neural Information Processing Systems, volume 33, pages 1747–1758. Curran Associates, Inc., 2020.
  21. Topology-preserving shape reconstruction and registration via neural diffeomorphic flow. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 20845–20855, 2022.
  22. A log-euclidean framework for statistics on diffeomorphisms. In Medical Image Computing and Computer-Assisted Intervention (MICCAI), pages 924–931. Springer, 2006.
  23. John Ashburner. A fast diffeomorphic image registration algorithm. Neuroimage, 38(1):95–113, 2007.
  24. Diffeomorphic demons: Efficient non-parametric image registration. NeuroImage, 45(1):S61–S72, 2009.
  25. Symmetric log-domain diffeomorphic registration: A demons-based approach. In Medical Image Computing and Computer-Assisted Intervention (MICCAI), pages 754–761. Springer, 2008.
  26. Computing large deformation metric mappings via geodesic flows of diffeomorphisms. International Journal of Computer Vision, 61(2):139–157, 2005.
  27. Surface matching via currents. In International Conference on Information Processing in Medical Imaging, pages 381–392. Springer, 2005.
  28. Geodesic shooting for computational anatomy. Journal of Mathematical Imaging and Vision, 24:209–228, 2006.
  29. Diffeomorphic 3d image registration via geodesic shooting using an efficient adjoint calculation. International Journal of Computer Vision, 97:229–241, 2012.
  30. Laurent Younes. Shapes and Diffeomorphisms. Springer, 2 edition, 2019.
  31. Fast diffeomorphic image registration via fourier-approximated lie algebras. International Journal of Computer Vision, 127:61–73, 2019.
  32. Unbiased diffeomorphic atlas construction for computational anatomy. NeuroImage, 23:S151–S160, 2004.
  33. Bayesian template estimation in computational anatomy. NeuroImage, 42(1):252–261, 2008.
  34. A bayesian generative model for surface template estimation. International Journal of Biomedical Imaging, 2010:1–14, 2010.
  35. Bayesian principal geodesic analysis for estimating intrinsic diffeomorphic image variability. Medical Image Analysis, 25(1):37–44, 2015.
  36. Probabilistic modeling of anatomical variability using a low dimensional parameterization of diffeomorphisms. Medical Image Analysis, 41:55–62, 2017.
  37. Implicit neural representations for deformable image registration. In International Conference on Medical Imaging with Deep Learning, pages 1349–1359. PMLR, 2022.
  38. Deformed implicit field: Modeling 3d shapes with learned dense correspondence. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 10286–10296, 2021.
  39. Deep implicit templates for 3d shape representation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 1429–1439, 2021.
  40. Pattern recognition and machine learning. Springer, 2006.
  41. Principal geodesic analysis for the study of nonlinear statistics of shape. IEEE Transactions on Medical Imaging, 23(8):995–1005, 2004.
  42. Probabilistic principal geodesic analysis. In Advances in Neural Information Processing Systems, volume 26, page 1178–1186. Curran Associates, Inc., 2013.
  43. Barycenters in the wasserstein space. SIAM Journal on Mathematical Analysis, 43(2):904–924, 2011.
  44. Principal geodesic analysis for probability measures under the optimal transport metric. In Advances in Neural Information Processing Systems, volume 28, pages 3312–3320. Curran Associates, Inc., 2015.
  45. Principal geodesic analysis in the space of discrete shells. Computer Graphics Forum, 37(5):173–184, 2018.
  46. Statistics on diffeomorphisms via tangent space representations. NeuroImage, 23:S161–S169, 2004.
  47. Distortion minimizing geodesic subspaces in shape spaces and computational anatomy. In VipIMAGE 2017, pages 1135–1144. Springer, 2018.
  48. Mark A Kramer. Nonlinear principal component analysis using autoassociative neural networks. AIChE journal, 37(2):233–243, 1991.
  49. Reducing the dimensionality of data with neural networks. Science, 313(5786):504–507, 2006.
  50. Learning weighted submanifolds with variational autoencoders and riemannian variational autoencoders. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 14503–14511, 2020.
  51. Latent space oddity: on the curvature of deep generative models. In International Conference on Learning Representations, 2018.
  52. Fast and robust shortest paths on manifolds learned from data. In Proceedings of the Twenty-Second International Conference on Artificial Intelligence and Statistics, volume 89 of Proceedings of Machine Learning Research, pages 1506–1515. PMLR, 2019.
  53. Metrics for deep generative models. In Proceedings of the Twenty-First International Conference on Artificial Intelligence and Statistics, volume 84 of Proceedings of Machine Learning Research, pages 1540–1550. PMLR, 2018.
  54. Geometry-aware hamiltonian variational auto-encoder. arXiv preprint arXiv:2010.11518, 2020.
  55. A prior-based approximate latent riemannian metric. In Proceedings of The 25th International Conference on Artificial Intelligence and Statistics, volume 151 of Proceedings of Machine Learning Research, pages 4634–4658. PMLR, 2022.
  56. Uniform interpolation constrained geodesic learning on data manifold. arXiv preprint arXiv:2002.04829, 2020.
  57. Geodesic clustering in deep generative models. arXiv preprint arXiv:1809.04747, 2018.
  58. Data augmentation with variational autoencoders and manifold sampling. In Deep Generative Models, and Data Augmentation, Labelling, and Imperfections, pages 184–192. Springer, 2021.
  59. Augmenting implicit neural shape representations with explicit deformation fields. arXiv preprint arXiv:2108.08931, 2021.
  60. Neural ordinary differential equations. In Advances in Neural Information Processing Systems, volume 31, page 6572–6583. Curran Associates, Inc., 2018.
  61. Hamiltonian neural networks. In Advances in Neural Information Processing Systems, volume 32, pages 15379–15389. Curran Associates, Inc., 2019.
  62. Symplectic ode-net: Learning hamiltonian dynamics with control. arXiv preprint arXiv:1909.12077, 2019.
  63. Lagrangian neural networks. arXiv preprint arXiv:2003.04630, 2020.
  64. TrajectoryNet: A dynamic optimal transport network for modeling cellular dynamics. In Proceedings of the 37th International Conference on Machine Learning, volume 119 of Proceedings of Machine Learning Research, pages 9526–9536. PMLR, 2020.
  65. A neural network approach for high-dimensional optimal control applied to multiagent path finding. IEEE Transactions on Control Systems Technology, 31(1):235–251, 2023.
  66. Discovering efficient periodic behaviors in mechanical systems via neural approximators. Optimal Control Applications and Methods, 44(6):3052–3079, 2023.
  67. Potential flow generator with l2 optimal transport regularity for generative models. IEEE Transactions on Neural Networks and Learning Systems, 33(2):528–538, 2022.
  68. Occupancy flow: 4d reconstruction by learning particle dynamics. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pages 5379–5389, 2019.
  69. A computational fluid mechanics solution to the monge-kantorovich mass transfer problem. Numerische Mathematik, 84(3):375–393, 2000.
  70. A machine learning framework for solving high-dimensional mean field game and mean field control problems. Proceedings of the National Academy of Sciences, 117(17):9183–9193, 2020.
  71. How to train your neural ODE: the world of Jacobian and kinetic regularization. In Proceedings of the 37th International Conference on Machine Learning, volume 119 of Proceedings of Machine Learning Research, pages 3154–3164. PMLR, 2020.
  72. Ot-flow: Fast and accurate continuous normalizing flows via optimal transport. Proceedings of the AAAI Conference on Artificial Intelligence, 35(10):9223–9232, 2021.
  73. Marching cubes: A high resolution 3d surface construction algorithm. SIGGRAPH Comput. Graph., 21(4):163–169, 1987.
  74. Deepsdf: Learning continuous signed distance functions for shape representation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 165–174, 2019.
  75. Implicit neural representations with periodic activation functions. In Advances in Neural Information Processing Systems, volume 33, pages 7462–7473. Curran Associates, Inc., 2020.
  76. Occupancy networks: Learning 3d reconstruction in function space. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 4460–4470, 2019.
  77. Overview of the geometries of shape spaces and diffeomorphism groups. Journal of Mathematical Imaging and Vision, 50:60–97, 2014.
  78. On the metrics and euler-lagrange equations of computational anatomy. Annual review of biomedical engineering, 4(1):375–405, 2002.
  79. A vector momenta formulation of diffeomorphisms for improved geodesic regression and atlas construction. In 2013 IEEE 10th International Symposium on Biomedical Imaging, pages 1219–1222. IEEE, 2013.
  80. Bayesian atlas building with hierarchical priors for subject-specific regularization. In Medical Image Computing and Computer Assisted Intervention (MICCAI), pages 76–86. Springer, 2021.
  81. Laurent Younes. Jacobi fields in groups of diffeomorphisms and applications. Quarterly of Applied Mathematics, 65(1):113–134, 2007.
  82. Computing an average anatomical atlas using lddmm and geodesic shooting. In 3rd IEEE International Symposium on Biomedical Imaging: Nano to Macro, pages 1116–1119. IEEE, 2006.
  83. Implicit geometric regularization for learning shapes. In International Conference on Machine Learning, pages 3789–3799. PMLR, 2020.
  84. As-killing-as-possible vector fields for planar deformation. Computer Graphics Forum, 30(5):1543–1552, 2011.
  85. Near-isometric level set tracking. Computer Graphics Forum, 35(5):65–77, 2016.
  86. A deep learning-based auto-segmentation system for organs-at-risk on whole-body computed tomography images for radiation therapy. Radiotherapy and Oncology, 160:175–184, 2021.
  87. Deformetrica 4: An open-source software for statistical shape analysis. In Shape in Medical Imaging, pages 3–13. Springer, 2018.
  88. Jürgen Jost. Riemannian geometry and geometric analysis. Springer, 7 edition, 2017.
Citations (1)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now