3D Point Cloud Classification Algorithm Based on Residual Edge Convolution
DU Zijin1, CAO Feilong1, YE Hailiang1, LIANG Jiye2
1. College of Sciences, China Jiliang University, Hangzhou 310018 2. Key Laboratory of Computational Intelligence and Chinese Information Processing of Ministry of Education, Shanxi University, Taiyuan 030006
Abstract The irregularity and disorder of 3D point cloud make the classification of point cloud more challenging. Therefore, a 3D point cloud classification algorithm based on residual edge convolution is designed. The discriminative shape descriptor can be learned from the point cloud directly and used for target classification. Firstly, an edge convolution block with residual learning is designed for feature extraction on the point cloud. In the edge convolution block, local graph is constructed with the input point cloud through the K-nearest neighbor algorithm and the local features are extracted and aggregated via convolution and maximum pooling, respectively. Subsequently, global features are extracted from the original point features through the multi-layer perceptron and combined with the local features in a residual learning way. Finally, a deep neural convolution network is constructed with the convolution block regarded as the basic unit to realize the classification of 3D point cloud. The organic combination of local features and global features is considered comprehensively. With a deeper structure, the network makes the final shape descriptor more abstract and discriminative. Experiments on two challenging datasets, ModelNet40 and ScanObjectNN, show that the proposed method obtains superior classification results.
Fund:National Natural Science Foundation of China(No.62032022,62006215,61876103)
Corresponding Authors:
CAO Feilong, Ph.D., professor. His research interests include deep learning and image processing.
About author:: DU Zijin, master student. His research interests include deep learning and point cloud analysis. YE Hailiang, Ph.D., lecturer. His research interests include deep learning and image processing. LIANG Jiye, Ph.D., professor. His research interests include artificial intelligence, granular computing and data mining.
[1] GUO Y L, WANG H Y, HU Q Y, et al. Deep Learning for 3D Point Clouds: A Survey. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020. DOI: 10.1109/TPAMI.2020.3005434. [2] RAHMAN M M, TAN Y H, XUE J, et al. Recent Advances in 3D Object Detection in the Era of Deep Neural Networks: A Survey. IEEE Transactions on Image Processing, 2019, 29: 2947-2962. [3] HUI Z Y, HU Y J, JIN S G, et al. Road Centerline Extraction from Airborne LiDAR Point Cloud Based on Hierarchical Fusion and Optimization. ISPRS Journal of Photogrammetry and Remote Sensing, 2016, 118: 22-36. [4] QI C R, SU H, NIEβNER M, et al. Volumetric and Multi-view CNNs for Object Classification on 3D Data // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2016: 5648-5656. [5] SU H, MAJI S, KALOGERAKIS E, et al. Multi-view Convolutional Neural Networks for 3D Shape Recognition // Proc of the IEEE International Conference on Computer Vision. Washington, USA: IEEE, 2015: 945-953. [6] QI C R, SU H, KAICHUN M, et al. PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2017: 77-85. [7] QI C R, YI L, SU H, et al. PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space // Proc of the 31st Conference on Neural Information Processing Systems. Cambridge, USA: The MIT Press, 2017: 5099-5108. [8] LEI H, AKHTAR N, MIAN A. Spherical Kernel for Efficient Graph Convolution on 3D Point Clouds. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020. DOI: 10.1109/TPAMI.2020.2983410. [9] FENG M T, ZHANG L, LIN X F, et al. Point Attention Network for Semantic Segmentation of 3D Point Clouds. Pattern Recognition, 2020, 107. DOI: 10.1016/j.patcoq.2020.107446. [10] CHEN C F, QIAN S S, FANG Q, et al. HAPGN: Hierarchical Attentive Pooling Graph Network for Point Cloud Segmentation. IEEE Transactions on Multimedia, 2020. DOI: 10.1109/TMM.2020.3009499. [11] SIMONOVSKY M, KOMODAKIS N. Dynamic Edge-Conditioned Filters in Convolutional Neural Networks on Graphs // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2017: 29-38. [12] WANG Y, SUN Y B, LIU Z W, et al. Dynamic Graph CNN for Learning on Point Clouds. ACM Transactions on Graphics, 2019, 38(5). DOI: 10.1145/3326362. [13] SUTSKEVER I, MARTENS J, DAHL G, et al. On the Importance of Initialization and Momentum in Deep Learning // Proc of the 30th International Conference on Machine Learning. New York, USA: ACM, 2013, III: 1139-1147. [14] MÜLLER R, KORNBLITH S, HINTON G. When Does Label Smoothing Help? // Proc of the 33rd Conference on Neural Information Processing Systems. Cambridge, USA: The MIT Press, 2019: 4694-4703. [15] LOSHCHILOV I, HUTTER F. SGDR: Stochastic Gradient Descent with Warm Restarts[C/OL]. [2021-01-15]. http://aad.informatik.uni-freiburg.de/papers/17-ICLR-SGDR.pdf. [16] WU Z R, SONG S R, KHOSLA A, et al. 3D ShapeNets: A Deep Representation for Volumetric Shapes // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2015: 1912-1920. [17] UY M A, PHAM Q H, HUA B S, et al. Revisiting Point Cloud Classification: A New Benchmark Dataset and Classification Model on Real-World Data // Proc of the IEEE/CVF International Conference on Computer Vision. Washington, USA: IEEE, 2019: 1588-1597. [18] XU Y F, FAN T Q, XU M Y, et al. SpiderCNN: Deep Learning on Point Sets with Parameterized Convolutional Filters // Proc of the European Conference on Computer Vision. Berlin, Germany: Springer, 2018: 90-105. [19] KLOKOV R, LEMPITSKY V. Escape from Cells: Deep KD-Networks for the Recognition of 3D Point Cloud Models // Proc of the IEEE International Conference on Computer Vision. Washington, USA: IEEE, 2017: 863-872. [20] LI J X, CHEN B M, LEE G H. SO-Net: Self-organizing Network for Point Cloud Analysis // Proc of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2018: 9397-9406. [21] LI Y Y, BU R, SUN M C, et al. PointCNN: Convolution on Χ-transformed Points // Proc of the 32nd Conference on Neural Information Processing Systems. Cambridge, USA: The MIT Press, 2018: 828-838. [22] WU W X, QI Z G, LI F X. PointConv: Deep Convolutional Networks on 3D Point Clouds // Proc of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2019: 9613-9622. [23] ATZMON M, MARON H, LIPMAN Y. Point Convolutional Neural Networks by Extension Operators. ACM Transactions on Graphics, 2018, 37(4). DOI: 10.1145/3197517.3201301. [24] KOMARICHEV A, ZHONG Z C, HUA J. A-CNN: Annularly Con-volutional Neural Networks on Point Clouds // Proc of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2019: 7421-7430. [25] LIU Y C, FAN B, XIANG S M, et al. Relation-Shape Convolutional Neural Network for Point Cloud Analysis // Proc of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2019: 8895-8904. [26] LAN S Y, YU R C, YU G, et al. Modeling Local Geometric Structure of 3D Point Clouds Using Geo-CNN // Proc of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2019: 998-1008. [27] BOULCH A. ConvPoint: Continuous Convolutions for Point Cloud Processing. Computers and Graphics, 2020, 88: 24-34. [28] YAN X, ZHENG C D, LI Z, et al. PointASNL: Robust Point Clouds Processing Using Nonlocal Neural Networks with Adaptive Sampling // Proc of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2020: 5589-5598. [29] BEN-SHABAT Y, LINDENBAUM M, FISCHER A. 3D Point Cloud Classification and Segmentation Using 3D Modified Fisher Vector Representation for Convolutional Neural Networks[C/OL]. [2021-01-15]. https://arxiv.org/pdf/1711.08241.pdf. [30] VAN DER MAATEN L, HINTON G. Visualizing Data Using T-SNE. Journal of Machine Learning Research, 2008, 9: 2579-2605.
2021, Vol. 34 
