Abstract
Appropriate weight initialization has been of key importance to successfully train neural networks. Recently, batch normalization has diminished the role of weight initialization by simply normalizing each layer based on batch statistics. Unfortunately, batch normalization has several drawbacks when applied to small batch sizes, as they are required to cope with memory limitations when learning on point clouds. While well-founded weight initialization strategies can render batch normalization unnecessary and thus avoid these drawbacks, no such approaches have been proposed for point convolutional networks. To fill this gap, we propose a framework to unify the multitude of continuous convolutions. This enables our main contribution, variance-aware weight initialization. We show that this initialization can avoid batch normalization while achieving similar and, in some cases, better performance.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Atzmon, M., Maron, H., Lipman, Y.: Point convolutional neural networks by extension operators. In: ACM Transactions on Graphics (Proc. SIGGRAPH) (2018)
Boulch, A.: ConvPoint: continuous convolutions for point cloud processing. Comput. Graph. 88, 24–34 (2020)
Choy, C., Gwak, J., Savarese, S.: 4D spatio-temporal convnets: minkowski convolutional neural networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 3075–3084 (2019)
Dai, A., Chang, A.X., Savva, M., Halber, M., Funkhouser, T., Nießner, M.: ScanNet: richly-annotated 3D reconstructions of indoor scenes. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 5828–5839 (2017)
Glorot, X., Bengio, Y.: Understanding the difficulty of training deep feedforward neural networks. In: Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics (2010)
Groh, F., Wieschollek, P., Lensch, H.P.A.: Flex-convolution (million-scale point-cloud learning beyond grid-worlds). In: ACCV (2018)
He, K., Zhang, X., Ren, S., Sun, J.: Delving deep into rectifiers: surpassing human-level performance on ImageNet classification. In: ICCV, pp. 1026–1034 (2015)
Hermosilla, P., Ritschel, T., Vazquez, P.P., Vinacua, A., Ropinski, T.: Monte Carlo convolution for learning on non-uniformly sampled point clouds. In: ACM Transactions on Graphics (Pro. of SIGGRAPH Asia) (2018)
Hua, B.S., Tran, M.K., Yeung, S.K.: Pointwise convolutional neural networks. In: Computer Vision and Pattern Recognition (CVPR) (2018)
Ioffe, S., Szegedy, C.: Batch normalization: accelerating deep network training by reducing internal covariate shift. In: International Conference on Machine Learning, pp. 448–456. PMLR (2015)
Krähenbühl, P., Doersch, C., Donahue, J., Darrell, T.: Data-dependent initializations of convolutional neural networks. arXiv:1511.06856 (2015)
Lei, H., Akhtar, N., Mian, A.: Octree guided CNN with spherical kernels for 3D point clouds. In: CVPR (2019)
Mao, J., Wang, X., Li, H.: Interpolated convolutional networks for 3D point cloud understanding. In: International Conference on Computer Vision (ICCV) (2019)
Mishkin, D., Matas, J.: All you need is a good init. arXiv:1511.06422 (2015)
Nekrasov, A., Schult, J., Litany, O., Leibe, B., Engelmann, F.: Mix3D: out-of-context data augmentation for 3D scenes. In: International Conference on 3D Vision (3DV) (2021)
Qi, C.R., Su, H., Mo, K., Guibas, L.J.: PointNet: deep learning on point sets for 3D classification and segmentation. In: CVPR, pp. 652–660 (2017)
Ronneberger, O., Fischer, P., Brox, T.: U-Net: convolutional networks for biomedical image segmentation. In: Navab, N., Hornegger, J., Wells, W.M., Frangi, A.F. (eds.) MICCAI 2015. LNCS, vol. 9351, pp. 234–241. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-24574-4_28
Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. arXiv:1409.1556 (2014)
Thomas, H., Qi, C.R., Deschaud, J.E., Marcotegui, B., Goulette, F., Guibas, L.J.: KPConv: flexible and deformable convolution for point clouds. In: ICCV (2019)
Uy, M.A., Pham, Q.H., Hua, B.S., Nguyen, D.T., Yeung, S.K.: Revisiting point cloud classification: A new benchmark dataset and classification model on real-world data. In: International Conference on Computer Vision (ICCV) (2019)
Wu, W., Qi, Z., Fuxin, L.: PointConv: deep convolutional networks on 3D point clouds. In: CVPR (2019)
Wu, Y., He, K.: Group normalization. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) ECCV 2018. LNCS, vol. 11217, pp. 3–19. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01261-8_1
Wu, Z., et al.: 3D shapeNets: a deep representation for volumetric shape modeling. In: Proceedings of 28th IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2015)
Acknowledgements
This work was partially funded by the Deutsche Forschungsgemeinschaft (DFG) under grant 391088465 (ProLint) and by the Federal Ministry of Health (BMG) under grant ZMVI1-2520DAT200 (AktiSmart-KI).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
1 Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Hermosilla, P., Schelling, M., Ritschel, T., Ropinski, T. (2022). Variance-Aware Weight Initialization for Point Convolutional Neural Networks. 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 13688. Springer, Cham. https://doi.org/10.1007/978-3-031-19815-1_5
Download citation
DOI: https://doi.org/10.1007/978-3-031-19815-1_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-19814-4
Online ISBN: 978-3-031-19815-1
eBook Packages: Computer ScienceComputer Science (R0)