Abstract
Point cloud super-resolution plays a central role in the mesh's quality in 3D reconstruction, while the feature extractor is vital for the learning-based point cloud upsampling pipelines. In this paper, we propose an arbitrary 3D point cloud upsampling network (PSR-GAT), which comprises the feature extraction module, GAT module, and upsampling module. For the input point cloud, the feature extraction module locates k nearest points of each point in 3D space by k-NN algorithm, then converts the local geometry information into high dimensional feature space through a multi-layer point-wise convolution. The GAT module converts the local geometry feature of each point into the semantic feature through a multi-layer graph attention network. The module dynamically adjusts the neighbor space of the point in each layer to increase the receptive field range and effectively fuses the semantic information of different levels through residual connection. This makes the local geometric in- formation extraction efficient. The upsampling module adds the number of points and maps them from feature space to 3D space. Extensive experimental results show that PSR-GAT exhibits a better performance than the existing state-of-the-art approaches.
Similar content being viewed by others
Data availability
All data included in this study are available upon request by contact with the corresponding author.
References
Alexa M, Behr J, Cohen-Or D, Fleishman S, Levin D, Silva CT (2003) Computing and rendering point set surfaces. IEEE Trans Visual Comput Graphics 9(1):3–15
Amirkhani D, Bastanfard A (2019) Inpainted image quality evaluation based on saliency map features. In: 2019 5th Iranian Conference on Signal Processing and Intelligent Systems (ICSPIS), pp. 1–6. IEEE
Bastanfard A, Amirkhani D, Mohammadi M (2022) Toward image super- resolution based on local regression and nonlocal means. Multimed Tools Appl 81(16):23473–23492
Boltz S, Nielsen F, Soatto S (2010) Earth mover distance on superpixels. In: International Conference on Image Processing, HongKong, China, pp. 4597–4600
Butt MA, Maragos P (1998) Optimum design of chamfer distance transforms. IEEE Trans Image Process 7(10):1477–1484
Chen TY, Hsiao CC, Huang C-C (2022) Density-imbalance-eased lidar point cloud upsampling via feature consistency learning. IEEE Trans Intell Veh 1–1. https://doi.org/10.1109/TIV.2022.3162672
Chen Y, Liu L, Phonevilay V, Gu K, Xia R, Xie J, Zhang Q, Yang K (2021) Image super-resolution reconstruction based on feature map attention mechanism. Appl Intell 51(7):4367–4380
Cuturi M (2013) Sinkhorn distances: Lightspeed computation of optimal transport. Advances in Neural Information Processing Systems 26.
Dong Y, Cordonnier J-B, Loukas A (2021) Attention is not all you need: Pure attention loses rank doubly exponentially with depth. In: International Conference on Machine Learning, pp. 2793–2803
Dong C, Loy CC, He K, Tang X (2015) Image super-resolution using deep convolutional networks. IEEE Trans Pattern Anal Mach Intell 38(2):295–307
Geiger A, Lenz P, Stiller C, Urtasun R (2013) Vision meets robotics: The KITTI dataset. Int J Robotics Res 32(11):1231–1237. https://doi.org/10.1177/0278364913491297
Geiger A, Lenz P, Urtasun R (2012) Are we ready for autonomous driving? the kitti vision benchmark suite. In: Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognition, Providence, RI, USA, pp. 3354–3361
Guennebaud G, Barthe L, Paulin M (2004) Real-time point cloud refinement. In: PBG, pp. 41–48
Guo MH, Cai JX, Liu ZN, Mu TJ, Martin RR, Hu SM (2021) Pct: Point cloud transformer. Comput Vis Media 7(2):187–199
Guo F, Ren Q, Tang J, Li Z (2022) Dilated multi-scale fusion for point cloud classification and segmentation. Multimed Tools Appl 81:6069–6090. https://doi.org/10.1007/s11042-021-11825-9
Han B, Zhang X, Ren S (2022) Pu-gacnet: Graph attention convolution net- work for point cloud upsampling. Image Vis Comput 104371
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognition, Las Vegas, NV, USA, pp. 770–778
Hu X, Mu H, Zhang X, Wang Z, Tan T, Sun J (2019) Meta-sr: A magnification-arbitrary network for super-resolution. In: Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognition, Long Beach, USA, pp. 1575–1584
Huang H, Wu S, Gong M, Cohen-Or D, Ascher U, Zhang H (2013) Edge- aware point set resampling. ACM Trans Graph (TOG) 32(1):1–12
Kingma DP, Ba J (2015) Adam: A method for stochastic optimization. In: 3rd International Conference on Learning Representations, San Diego, CA, USA. http://arxiv.org/abs/1412.6980
Kipf TN, Welling M (2016) Semi-supervised classification with graph convolutional networks. arXiv preprint arXiv:1609.02907
Li D, He K, Wang L, Zhang D (2022) Local feature extraction network with high correspondences for 3d point cloud registration. Appl Intell 52:9638–9649. https://doi.org/10.1007/s10489-021-03055-1
Li R, Li X, Fu C-W, Cohen-Or D, Heng P-A (2019) Pu-gan: a point cloud upsampling adversarial network. In: Proc. IEEE/CVF International Conference on Computer Vision, Long Beach, CA, USA, pp. 7203–7212
Li R, Li X, Heng P, Fu C (2021) Point cloud upsampling via disentangled refinement. In: Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognition, pp. 344–353. https://openaccess.thecvf.com/content/CVPR2021/html/Li_Point_Cloud_Upsampling_via_Disentangled_Refinement_CVPR_2021_paper.html
Li R, Li X, Heng P-A, Fu C-W (2021) Point cloud upsampling via dis- entangled refinement. In: Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognition, pp. 344–353
Lipman Y, Cohen-Or D, Levin D, Tal-Ezer H (2007) Parameterization-free projection for geometry reconstruction. ACM Trans Graph (TOG) 26(3):22
Luo C, Chen Y, Wang N, Zhang Z (2019) Spectral feature transformation for person re-identification. In: Proc. IEEE/CVF International Conference on Computer Vision, Seoul, Korea, pp. 4976–4985
Luo L, Tang L, Zhou W, Wang S, Yang Z (2021) PU-EVA: an edge-vector based approximation solution for flexible-scale point cloud upsampling. In: Proc. International Conference on Computer Vision, pp. 16188–16197. https://doi.org/10.1109/ICCV48922.2021.01590
Mansourifar H, Bastanfard A (2011) A novel practical approach for weight manipulation of conic splines. In: 2011 Eighth International Conference Computer Graphics, Imaging and Visualization, pp. 65–69. IEEE
Mansourifar H, Bastanfard A (2011) Fast conic spline data fitting of noise- free data points. In: 2011 Eighth International Conference Computer Graphics, Imaging and Visualization, pp. 59–64. IEEE
Mao J, Xue Y, Niu M, Bai H, Feng J, Liang X, Xu H, Xu C (2021) Voxel transformer for 3d object detection. In: Proc. IEEE/CVF International Conference on Computer Vision, pp. 3164–3173
Maturana D, Scherer S (2015) Voxnet: A 3d convolutional neural network for real-time object recognition. In: Proc. IEEE/RSJ International Conference on Intelligent Robots and Systems, Hamburg, Germany, pp. 922–928
Mei Y, Fan Y, Zhou Y (2021) Image super-resolution with non-local sparse attention. In: Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognition, pp. 3517–3526. https://openaccess.thecvf.com/content/CVPR2021/html/MeiImageSuper-ResolutionWithNon-LocalSparseAttentionCVPR2021paper.html
Paszke A, Gross S, Massa F, Lerer A, Bradbury J, Chanan G, Killeen T, Lin Z, Gimelshein N, Antiga L, Desmaison A, Kopf A, Yang EZ, DeVito Z, Raison M, Tejani A, Chilamkurthy S, Steiner B, Fang L, Bai J, Chintala S (2019) Pytorch: An imperative style, high-performance deep learning library. In: Advances in Neural Information Processing Systems, Vancouver, Canada, pp. 8024–8035. https://proceedings.neurips.cc/paper_files/paper/2019/file/bdbca288fee7f92f2bfa9f7012727740-Paper.pdf. Accessed 14 Dec 2019
Qi CR, Su H, Mo K, Guibas LJ (2017) Pointnet: Deep learning on point sets for 3d classification and segmentation. In: Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognition, Honolulu, HI, USA, pp. 652–660
Qi CR, Yi L, Su H, Guibas LJ (2017) Pointnet++: Deep hierarchical feature learning on point sets in a metric space. Adv Neural Inf Process Syst 30
Qian G, Abualshour A, Li G, Thabet A, Ghanem B (2021) Pu-gcn: Point cloud upsampling using graph convolutional networks. In: Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognition, pp. 11683–11692
Qian Y, Hou J, Kwong S, He Y (2020) Pugeo-net: A geometry-centric net- work for 3d point cloud upsampling. In: Proc. European Conference of Computer Vision. Lecture Notes in Computer Science, vol. 12364, pp. 752–769. https://doi.org/10.1007/978-3-030-58529-7_44
Qiao C, Li D, Guo Y, Liu C, Jiang T, Dai Q, Li D (2021) Evaluation and development of deep neural networks for image super-resolution in optical microscopy. Nat Methods 18(2):194–202
Simonovsky M, Komodakis N (2017) Dynamic edge-conditioned filters in convolutional neural networks on graphs. In: Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognition, Honolulu, HI, USA, pp. 3693–3702
Sokolova M, Japkowicz N, Szpakowicz S (2006) Beyond accuracy, f-score and roc: a family of discriminant measures for performance evaluation. In: Australasian Joint Conference on Artificial Intelligence, Hobart, Australasian, pp. 1015–1021
Song W, Xiao Z, Wang Y, Charlin L, Zhang M, Tang J (2019) Session- based social recommendation via dynamic graph attention networks. In: Proc. ACM International Conference on Web Search and Data Mining, Melbourne, Australia, pp. 555–563. https://doi.org/10.1145/3289600.3290989
Tang H, Liu Z, Zhao S, Lin Y, Lin J, Wang H, Han S (2020) Searching efficient 3d architectures with sparse point-voxel convolution. In: Vedaldi A, Bischof H, Brox T, Frahm J (eds.) Proc. European Conference of Computer Vision. Lecture Notes in Computer Science, vol. 12373, pp. 685–702. https://doi.org/10.1007/978-3-030-58604-1_41
Te G, Hu W, Zheng A, Guo Z (2018) Rgcnn: Regularized graph cnn for point cloud segmentation. In: Proc. 26th ACM International Conference on Multimedia, Seoul, Korea, pp. 746–754
van Erven T, Harremoes P (2014) Renyi divergence and kullback-leibler divergence. IEEE Trans Inf Theory 60(7):3797–3820. https://doi.org/10.1109/TIT.2014.2320500
Velickovic P, Cucurull G, Casanova A, Romero A, Lio P, Bengio Y (2017) Graph attention networks. arXiv preprint arXiv:1710.10903
Wang H, Liu X, Kang W, Yan Z, Wang B, Ning Q (2022) Multi-features guidance network for partial-to-partial point cloud registration. Neural Comput Appl 34:1623–1634. https://doi.org/10.1007/s00521-021-06464-y
Wang G, Xu G, Wu Q, Wu X (2021) Two-stage point cloud super resolution with local interpolation and readjustment via outer-product neural network. J Syst Sci Complexity 34(1):68–82
Wu H, Zhang J, Huang K (2019) Point cloud super resolution with adver- sarial residual graph networks. arXiv preprint arXiv:1908.02111
Xu J, Le X, Chen C, Guan X (2022) Spinet: self-supervised point cloud frame interpolation network. Neural Comput Appl. https://doi.org/10.1007/s00521-022-06939-6
Yang J, Wright J, Huang TS, Ma Y (2010) Image super-resolution via sparse representation. IEEE Trans Image Process 19(11):2861–2873
Ye S, Chen D, Han S, Wan Z, Liao J (2021) Meta-pu: An arbitrary-scale upsampling network for point cloud. IEEE Trans Vis Comput Graphics 28(9):3206–3218
Yu L, Li X, Fu C, Cohen Or D, Heng P (2018) Pu-net: Point cloud upsampling network. In: Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognition, Salt Lake City, UT, USA, pp. 2790–2799
Yu X, Rao Y, Wang Z, Liu Z, Lu J, Zhou J (2021) Pointr: Diverse point cloud completion with geometry-aware transformers. In: Proc. IEEE/CVF International Conference on Computer Vision, pp. 12498–12507
Zhang Z (2012) Microsoft kinect sensor and its effect. IEEE Multim 19(2):4–10. https://doi.org/10.1109/MMUL.2012.24
Zhang W, Fu X, Li W (2020) 3d modeling system of lidar point cloud pro- cessing algorithm based on artificial intelligence. In: Proc. Applications and Techniques in Cyber Intelligence. Advances in Intelligent Systems and Computing, vol. 1244, pp. 764–770. Springer, Fuyang, China
Zhang F, Nauata N, Furukawa Y (2020) Conv-mpn: Convolutional message passing neural network for structured outdoor architecture reconstruction. In: Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognition, Seattle, WA, USA, pp. 2795–2804. https://doi.org/10.1109/CVPR42600.2020.00287
Zhao Y, Hui L, Xie J (2021) Sspu-net: Self-supervised point cloud upsampling via differentiable rendering. In: Proc. Multimedia Conference, pp. 2214–2223. https://doi.org/10.1145/3474085.3475381
Zolanvari SMI, Ruano S, Rana A, Cummins A, da Silva RE, Rahbar M, Smolic A (2019) Dublincity: Annotated lidar point cloud and its applications. In: Proc. British Machine Vision Conferenc, Cardiff, UK, p. 44. https://bmvc2019.org/wp-content/uploads/papers/0644-paper.pdf. Accessed 2019-09-06
Acknowledgements
This work was supported by the Yunnan Major Scien- tific and Technological Special Project under Grant 202002AD080001.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflicts of interests/Competing interests
We declare that we do not have any commercial or associative interest that represents a conflict of interest in connection with the work submitted.
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Zhong, F., Bai, Z. PSR-GAT: Arbitrary point cloud super-resolution using graph attention networks. Multimed Tools Appl 83, 26213–26232 (2024). https://doi.org/10.1007/s11042-023-16525-0
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-023-16525-0