Abstract
The sparsity and incompleteness of point clouds generally result in challenges in point cloud analysis. Most existing point cloud completion methods use an individual Euclidean space feature to generate point clouds. Consequently, the generated point clouds are relatively rough. This paper proposes a multi-space and detail-supplemented attention point cloud completion network (MSDSA-Net). Here, the key is to utilize multi-space features to generate high-quality point clouds. First, we construct a dual-branch multi-space feature extractor (MSFE). A branch of the MSFE is a local-holistic geometric feature extractor based on Euclidean space and eigenvalue space. It can extract features with similar local geometric structures at points that are at a distance, to compensate for the missing part of the feature information of the point cloud. Another branch of the MSFE is a global feature extractor based on Euclidean space to extract the global features. Second, we continue to follow the coarse-to-fine decoding framework of general completion networks. However, in the fine generation stage, we propose a detail-supplemented (DS) module to supplement the features used to guide point cloud generation in detail. Extensive experiments demonstrate that our network has a good effect on the shape completion of point clouds.
Similar content being viewed by others
References
Ratnasingam S (2019) Deep camera: a fully convolutional neural network for image signal processing. In: Proceedings of 2019 IEEE/CVF international conference on computer vision workshop (ICCVW), IEEE, Seoul, Korea (South), pp 3868–3878
Zabatani A, Surazhsky V, Sperling E, Moshe SB, Menashe O, Silver DH, Karni Z, Bronstein AM, Bronstein MM, Kimmel R (2019) Intel®; realsenseTM sr300 coded light depth camera. IEEE Trans Pattern Anal Mach Intell 42(10):2333–2345
Li W, Wang FD, Xia GS (2020) A geometry-attentional network for ALS point cloud classification. ISPRS J Photogramm Remote Sensing 164:26–40
Zhang M, You H, KP LS, Kuo CCJ (2020) Pointhop: an explainable machine learning method for point cloud classification. IEEE Trans Multimedia 22(7):1744–1755
Lin Y, Vosselman G, Cao Y, Yang MY (2020) Active and incremental learning for semantic ALS point cloud segmentation. ISPRS J Photogramm Remote Sens 169:73–92
Xie Y, Tian J, Zhu XX (2020) Linking points with labels in 3D: a review of point cloud semantic segmentation. IEEE Geosci Remote Sensing Magazine 8(4):38–59
Rahman MM, Tan Y, Xue J, Lu K (2019) Recent advances in 3D object detection in the era of deep neural networks: a survey. IEEE Trans Image Process 29:2947–2962
Zhao ZQ, Zheng P, St X, Wu X (2019) Object detection with deep learning: a review. IEEE Trans Neural Netw Learn Syst 30(11):3212–3232
Feng M, Zhang L, Lin X, Gilani SZ, Mian A (2020) Point attention network for semantic segmentation of 3d point clouds. Pattern Recogn 107:107446
Jhaldiyal A, Chaudhary N (2022) Semantic segmentation of 3d lidar data using deep learning: a review of projection-based methods. Appl Intell:1–12
Yuan W, Khot T, Held D, Mertz C, Hebert M (2018) PCN: point completion network. In: Proceedings of the 2018 International Conference on 3D Vision, IEEE, Verona, Italy, pp 728–737
Tchapmi LP, Kosaraju V, Rezatofighi H, Reid I, Savarese S (2019) TopNet: structural point cloud decoder. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, long beach, USA, pp 383–392
Wang X, Ang Jr MH, Lee GH (2020) Cascaded refinement network for point cloud completion. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp 790–799
Pan L, Chen X, Cai Z, Zhang J, Zhao H, Yi S, Liu Z (2021) Variational relational point completion network. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, Nashville, USA, pp 8520–8529
Wen X, Li T, Han Z, Liu YS (2020) Point cloud completion by skip-attention network with hierarchical folding. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, Seattle, USA, pp 1939–1948
Qi CR, Su H, Mo K, Guibas LJ (2017) PointNet: deep learning on point sets for 3d classification and segmentation. In: Proceedings of the IEEE conference on computer vision and pattern recognition, Honolulu, USA, pp 77–85
Qi CR, Yi L, Su H, Guibas LJ (2017) PointNet++: deep hierarchical feature learning on point sets in a metric space. In: Proceedings of the advances in neural information processing systems, Long Beach, USA, pp 5099–5108
Liu M, Sheng L, Yang S, Shao J, Hu SM (2020) Morphing and sampling network for dense point cloud completion. In: Proceedings of the AAAI conference on artificial intelligence, New York, USA, vol 34, pp 11596–11603
Wu H, Miao Y, Fu R (2021) Point cloud completion using multiscale feature fusion and cross-regional attention. Image Vis Comput 111:104193
Xiang P, Wen X, Liu YS, Cao YP, Wan P, Zheng W, Han Z (2021) SnowflakeNet: point cloud completion by snowflake point deconvolution with skip-transformer. In: Proceedings of the IEEE/CVF international conference on computer vision, pp 5499–5509
Xie H, Yao H, Zhou S, Mao J, Zhang S, Sun W (2020) GRNEt: gridding residual network for dense point cloud completion. In: Proceedings of the european conference on computer vision. Springer, Newcastle University, UK, pp 365–381
Zhang W, Yan Q, Xiao C (2020) Detail preserved point cloud completion via separated feature aggregation. In: Proceedings of the european conference on computer vision. Springer, Newcastle University, UK, pp 512–528
Stutz D, Geiger A (2020) Learning 3d shape completion under weak supervision. Int J Comput Vis 128(5):1162–1181
Yang Y, Feng C, Shen Y, Tian D (2018) FoldingNet: point cloud auto-encoder via deep grid deformation. In: Proceedings of the IEEE conference on computer vision and pattern recognition, Salt Lake City, USA, pp 206–215
Zhang J, Chen W, Wang Y, Vasudevan R, Johnson-Roberson M (2021) Point set voting for partial point cloud analysis. IEEE Robot Autom Lett 6(2):596–603
Wen X, Xiang P, Han Z, Cao YP, Wan P, Zheng W, Liu YS (2021) PMP-Net: point cloud completion by learning multi-step point moving paths. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp 7443–7452
Guo MH, Xu TX, Liu JJ, Liu ZN, Jiang PT, Mu TJ, Zhang SH, Martin RR, Cheng MM, Hu SM (2022) Attention mechanisms in computer vision: a survey. Computat Vis Med:1–38
Mohammdi Farsani R, Pazouki E (2021) A transformer self-attention model for time series forecasting. J Electr Comput Eng Innovations (JECEI) 9(1):1–10
Niu Z, Zhong G, Yu H (2021) A review on the attention mechanism of deep learning. Neurocomput 452:48–62
Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, Kaiser Ł, Polosukhin I (2017) Attention is all you need. Adv Neural Inform Process Syst, vol 30
Fan H, Su H, Guibas LJ (2017) A point set generation network for 3d object reconstruction from a single image. In: Proceedings of the IEEE conference on computer vision and pattern recognition, Hawaii, America, pp 605–613
Tatarchenko M, Richter SR, Ranftl R, Li Z, Koltun V, Brox T (2019) What do single-view 3d reconstruction networks learn?. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, Los Angeles, USA, pp 3405–3414
Wang X, Ang MH, Lee GH (2020) Point cloud completion by learning shape priors. In: Proceedings of 2020 IEEE/RSJ international conference on intelligent robots and systems (IROS). IEEE, Las Vegas, USA, pp 10719–10726
Geiger A, Lenz P, Stiller C, Urtasun R (2013) Vision meets robotics: the KITTI dataset. Int J Robot Res 32(11):1231–1237
Kingma DP, Ba J (2015) Adam: a method for stochastic optimization. In: Proceedings of international conference on learning representations, San Diego, USA
Acknowledgements
This work was supported by the National Natural Science Foundation of China under grants 62032022, 62176244, and 62006215, and the Natural Science Foundation of Zhejiang (CN) under grants LZ20F030001 and LQ20F030016.
Author information
Authors and Affiliations
Corresponding author
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 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
Xiang, M., Ye, H., Yang, B. et al. Multi-space and detail-supplemented attention network for point cloud completion. Appl Intell 53, 14971–14985 (2023). https://doi.org/10.1007/s10489-022-04219-3
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10489-022-04219-3