Abstract
Point cloud semantic segmentation is a crucial problem in computer vision, which aims to assign semantic labels to each point in a point cloud. However, the sparsity and irregularity of point cloud data pose significant challenges to achieving accurate segmentation. Unlike traditional image classification tasks, each point in a point cloud not only has location information but also other feature information that must be considered. To address this issue, this paper proposes a novel point cloud semantic segmentation algorithm called PointAF. The proposed method utilizes soft projection operation during downsampling to better combine neighborhood information, an attention mechanism to achieve an adaptive offset effect, and residual connections to solve the problem of gradient disappearance. Experimental results show that the proposed method achieves great performance, with an mIoU of 70.6% and OA of 90.2% on the S3SDIS dataset, as well as mIoU of 69.2% and mACC of 70.1% on the ScanNetV2 dataset. The proposed method demonstrates great potential in point cloud semantic segmentation and may have practical applications in areas such as autonomous driving, robot navigation, and augmented reality.
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
References
Choy, C., Gwak, J., Savarese, S.: 4D spatio-temporal convnets: Minkowski convolutional neural networks. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 3075–3084 (2019)
Graham, B., Engelcke, M., Van Der Maaten, L.: 3D semantic segmentation with submanifold sparse convolutional networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 9224–9232 (2018)
Graham, B., Van der Maaten, L.: Submanifold sparse convolutional networks. arXiv preprint arXiv:1706.01307 (2017)
Qi, C.R., Su, H., Mo, K., Guibas, L.J.: PointNet: deep learning on point sets for 3D classification and segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 652–660 (2017)
Qi, C.R., Yi, L., Su, H., Guibas, L.J.: PointNet++: deep hierarchical feature learning on point sets in a metric space. In: Advances in Neural Information Processing Systems, 30 (2017)
Li, Y., Bu, R., Sun, M., Wu, W., Di, X., Chen, B.: PointCNN: convolution on x-transformed points. In: Advances in Neural Information Processing Systems, 31 (2018)
Thomas, H., Qi, C.R., Deschaud, J.E., Marcotegui, B., Goulette, F., Guibas, L.J.: KPConv: flexible and deformable convolution for point clouds. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 6411–6420 (2019)
Zhou, Y., Tuzel, O.: VoxelNet: end-to-end learning for point cloud based 3D object detection. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 4490–4499 (2018)
Riegler, G., Osman Ulusoy, A., Geiger, A.: OctNet: learning deep 3D representations at high resolutions. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 3577–3586 (2017)
Ochmann, S., Vock, R., Klein, R.: Automatic reconstruction of fully volumetric 3D building models from oriented point clouds. ISPRS J. Photogramm. Remote Sens. 151, 251–262 (2019)
Chen, X., Ma, H., Wan, J., Li, B., Xia, T.: Multi-view 3D object detection network for autonomous driving. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1907–1915 (2017)
Dai, A., Nießner, M.: 3DMV: joint 3D-multi-view prediction for 3D semantic scene segmentation. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) ECCV 2018. LNCS, vol. 11214, pp. 458–474. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01249-6_28
Gao, Z., Wang, D.Y., Xue, Y.B., Xu, G.P., Zhang, H., Wang, Y.L.: 3D object recognition based on pairwise multi-view convolutional neural networks. J. Vis. Commun. Image Represent. 56, 305–315 (2018)
Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. Commun. ACM 60(6), 84–90 (2017)
Vaswani, A., et al.: Attention is all you need. In: Advances in Neural Information Processing Systems, 30 (2017)
Phan, A.V., Le Nguyen, M., Nguyen, Y.L.H., Bui, L.T.: DGCNN: a convolutional neural network over large-scale labeled graphs. Neural Netw. 108, 533–543 (2018)
Zhao, H., Jiang, L., Jia, J., Torr, P.H., Koltun, V.: Point transformer. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 16259–16268 (2021)
Bahdanau, D., Cho, K., Bengio, Y.: Neural machine translation by jointly learning to align and translate. arXiv preprint arXiv:1409.0473 (2014)
Hu, Q., et al.: RandLA-Net: efficient semantic segmentation of large-scale point clouds. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 11108–11117 (2020)
Lai, X., et al.: Stratified transformer for 3D point cloud segmentation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 8500–8509 (2022)
Lang, I., Manor, A., Avidan, S.: SampleNet: differentiable point cloud sampling. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 7578–7588 (2020)
Jang, E., Gu, S., Poole, B.: Categorical reparameterization with gumbel-softmax. arXiv preprint arXiv:1611.01144 (2016)
Chen, R., Yan, X., Wang, S., Xiao, G.: DA-Net: dual-attention network for multivariate time series classification. Inf. Sci. 610, 472–487 (2022)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770–778 (2016)
Landrieu, L., Simonovsky, M.: Large-scale point cloud semantic segmentation with superpoint graphs. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 4558–4567 (2018)
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)
Acknowledgements
The work is supported by the National Key Research and Development Program of China (2021YFF0500903, 2022YFE0198900), the National Natural Science Foundation of China (52178271, 52077213).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Chen, T., Wang, X., Li, D., Liu, J., Wu, Z. (2023). PointAF: A Novel Semantic Segmentation Network for Point Cloud. In: Zhang, H., et al. International Conference on Neural Computing for Advanced Applications. NCAA 2023. Communications in Computer and Information Science, vol 1869. Springer, Singapore. https://doi.org/10.1007/978-981-99-5844-3_39
Download citation
DOI: https://doi.org/10.1007/978-981-99-5844-3_39
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-99-5843-6
Online ISBN: 978-981-99-5844-3
eBook Packages: Computer ScienceComputer Science (R0)