@article{Zhang_Song_Yao_Cai_2020, title={Shape-Oriented Convolution Neural Network for Point Cloud Analysis}, volume={34}, url={https://ojs.aaai.org/index.php/AAAI/article/view/6972}, DOI={10.1609/aaai.v34i07.6972}, abstractNote={<p>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 <em>ShapeConv</em> is proposed to focus on the representation learning of the underlying shape formed by each local neighboring point. Despite this <em>intra-shape relationship</em> learning, <em>ShapeConv</em> is also designed to incorporate the contextual effects from the <em>inter-shape relationship</em> 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.</p>}, number={07}, journal={Proceedings of the AAAI Conference on Artificial Intelligence}, author={Zhang, Chaoyi and Song, Yang and Yao, Lina and Cai, Weidong}, year={2020}, month={Apr.}, pages={12773-12780} }