Shape-Oriented Convolution Neural Network for Point Cloud Analysis


  • Chaoyi Zhang University of Sydney
  • Yang Song University of New South Wales
  • Lina Yao University of New South Wales
  • Weidong Cai University of Sydney



Point cloud is a principal data structure adopted for 3D geometric information encoding. Unlike other conventional visual data, such as images and videos, these irregular points describe the complex shape features of 3D objects, which makes shape feature learning an essential component of point cloud analysis. To this end, a shape-oriented message passing scheme dubbed ShapeConv is proposed to focus on the representation learning of the underlying shape formed by each local neighboring point. Despite this intra-shape relationship learning, ShapeConv is also designed to incorporate the contextual effects from the inter-shape relationship through capturing the long-ranged dependencies between local underlying shapes. This shape-oriented operator is stacked into our hierarchical learning architecture, namely Shape-Oriented Convolutional Neural Network (SOCNN), developed for point cloud analysis. Extensive experiments have been performed to evaluate its significance in the tasks of point cloud classification and part segmentation.




How to Cite

Zhang, C., Song, Y., Yao, L., & Cai, W. (2020). Shape-Oriented Convolution Neural Network for Point Cloud Analysis. Proceedings of the AAAI Conference on Artificial Intelligence, 34(07), 12773-12780.



AAAI Technical Track: Vision