Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
126 tokens/sec
GPT-4o
47 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

On the Fourier analysis in the SO(3) space : EquiLoPO Network (2404.15979v1)

Published 24 Apr 2024 in cs.CV, cs.LG, and math.GR

Abstract: Analyzing volumetric data with rotational invariance or equivariance is an active topic in current research. Existing deep-learning approaches utilize either group convolutional networks limited to discrete rotations or steerable convolutional networks with constrained filter structures. This work proposes a novel equivariant neural network architecture that achieves analytical Equivariance to Local Pattern Orientation on the continuous SO(3) group while allowing unconstrained trainable filters - EquiLoPO Network. Our key innovations are a group convolutional operation leveraging irreducible representations as the Fourier basis and a local activation function in the SO(3) space that provides a well-defined mapping from input to output functions, preserving equivariance. By integrating these operations into a ResNet-style architecture, we propose a model that overcomes the limitations of prior methods. A comprehensive evaluation on diverse 3D medical imaging datasets from MedMNIST3D demonstrates the effectiveness of our approach, which consistently outperforms state of the art. This work suggests the benefits of true rotational equivariance on SO(3) and flexible unconstrained filters enabled by the local activation function, providing a flexible framework for equivariant deep learning on volumetric data with potential applications across domains. Our code is publicly available at \url{https://gricad-gitlab.univ-grenoble-alpes.fr/GruLab/ILPO/-/tree/main/EquiLoPO}.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (38)
  1. On polynomial approximations for privacy-preserving and verifiable relu networks. arXiv preprint arXiv:2011.05530, 2020.
  2. Roto-translation covariant convolutional networks for medical image analysis. In Medical Image Computing and Computer Assisted Intervention–MICCAI 2018: 21st International Conference, Granada, Spain, September 16-20, 2018, Proceedings, Part I, pages 440–448. Springer, 2018.
  3. Group equivariant convolutional networks. In International conference on machine learning, pages 2990–2999. PMLR, 2016a.
  4. Steerable cnns. arXiv preprint arXiv:1612.08498, 2016b.
  5. Spherical cnns. arXiv preprint arXiv:1801.10130, 2018.
  6. Automatic symmetry discovery with lie algebra convolutional network. Advances in Neural Information Processing Systems, 34:2503–2515, 2021.
  7. A hitchhiker’s guide to geometric gnns for 3d atomic systems. arXiv preprint arXiv:2312.07511, 2023.
  8. Auto-sklearn: Efficient and robust automated machine learning. In Automated Machine Learning, pages 113–134. Springer International Publishing, 2019. doi: 10.1007/978-3-030-05318-5_6. URL https://doi.org/10.1007/978-3-030-05318-5_6.
  9. Vikas Gottemukkula. Polynomial activation functions. 2019.
  10. Deep residual learning for image recognition. In 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE, June 2016. doi: 10.1109/cvpr.2016.90. URL https://doi.org/10.1109/cvpr.2016.90.
  11. Spherical convolutions on molecular graphs for protein model quality assessment. Machine Learning: Science and Technology, 2(4):045005, 2021.
  12. Auto-keras: An efficient neural architecture search system. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. ACM, July 2019. doi: 10.1145/3292500.3330648. URL https://doi.org/10.1145/3292500.3330648.
  13. Highly accurate protein structure prediction with AlphaFold. Nature, 596(7873):583–589, July 2021. doi: 10.1038/s41586-021-03819-2. URL https://doi.org/10.1038/s41586-021-03819-2.
  14. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980, 2014.
  15. Exploiting redundancy: Separable group convolutional networks on lie groups. In International Conference on Machine Learning, pages 11359–11386. PMLR, 2022.
  16. Risi Kondor. N-body networks: a covariant hierarchical neural network architecture for learning atomic potentials. ArXiv, abs/1803.01588, 2018. URL https://api.semanticscholar.org/CorpusID:3665386.
  17. Clebsch-gordan nets: a fully fourier space spherical convolutional neural network, 2018.
  18. Imagenet classification with deep convolutional neural networks. Advances in neural information processing systems, 25, 2012.
  19. Regular se (3) group convolutions for volumetric medical image analysis. arXiv preprint arXiv:2306.13960, 2023.
  20. Multilayer feedforward networks with a nonpolynomial activation function can approximate any function. Neural Networks, 6(6):861–867, 1993. ISSN 0893-6080. doi: https://doi.org/10.1016/S0893-6080(05)80131-5. URL https://www.sciencedirect.com/science/article/pii/S0893608005801315.
  21. Feature pyramid vision transformer for MedMNIST classification decathlon. In 2022 International Joint Conference on Neural Networks (IJCNN). IEEE, July 2022a. doi: 10.1109/ijcnn55064.2022.9892282. URL https://doi.org/10.1109/ijcnn55064.2022.9892282.
  22. Group convolutional neural networks for dwi segmentation. In Geometric Deep Learning in Medical Image Analysis, pages 96–106. PMLR, 2022b.
  23. Protein model quality assessment using 3D oriented convolutional neural networks. Bioinformatics, 35(18):3313–3319, February 2019. doi: 10.1093/bioinformatics/btz122. URL https://doi.org/10.1093/bioinformatics/btz122.
  24. Attentive group equivariant convolutional networks. In International Conference on Machine Learning, pages 8188–8199. PMLR, 2020.
  25. Group convolutional neural networks improve quantum state accuracy. arXiv preprint arXiv:2104.05085, 2021.
  26. Clifford group equivariant neural networks. arXiv preprint arXiv:2305.11141, 2023.
  27. Dynamic routing between capsules. Advances in neural information processing systems, 30, 2017.
  28. Moving frame net: SE(3)-equivariant network for volumes. In NeurIPS Workshop on Symmetry and Geometry in Neural Representations, pages 81–97. PMLR, 2023.
  29. Tensor field networks: Rotation- and translation-equivariant neural networks for 3D point clouds. ArXiv, abs/1802.08219, 2018. URL https://api.semanticscholar.org/CorpusID:3457605.
  30. Deeply supervised 3D fully convolutional networks with group dilated convolution for automatic MRI prostate segmentation. Medical physics, 46(4):1707–1718, 2019.
  31. 3D steerable CNNs: Learning rotationally equivariant features in volumetric data. ArXiv, abs/1807.02547, 2018. URL https://api.semanticscholar.org/CorpusID:49654682.
  32. Maurice Weiler et al. Equivariant and coordinate independent convolutional networks: A gauge field theory of neural networks. 2024.
  33. 3D G-CNNs for pulmonary nodule detection. arXiv preprint arXiv:1804.04656, 2018.
  34. CubeNet: Equivariance to 3D rotation and translation. In Proceedings of the European Conference on Computer Vision (ECCV), pages 567–584, 2018.
  35. Reinventing 2D convolutions for 3D images. IEEE Journal of Biomedical and Health Informatics, 25(8):3009–3018, August 2021. doi: 10.1109/jbhi.2021.3049452. URL https://doi.org/10.1109/jbhi.2021.3049452.
  36. MedMNIST v2 - a large-scale lightweight benchmark for 2D and 3D biomedical image classification. Scientific Data, 10(1), January 2023. doi: 10.1038/s41597-022-01721-8. URL https://doi.org/10.1038/s41597-022-01721-8.
  37. ILPO-NET: Network for the invariant recognition of arbitrary volumetric patterns in 3d, 2024.
  38. 6DCNN with roto-translational convolution filters for volumetric data processing. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 36, pages 4707–4715, 2022.

Summary

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

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