DANC-Net: Dual-Attention and Negative Constraint Network for Point Cloud Classification

Abstract

Convolutional neural networks, as a branch of deep neural networks, have been widely used in multidimensional signal processing, especially in point cloud signal processing. Nevertheless, in point cloud signal processing, most point cloud classification networks currently do not consider local feature correlation. In addition, they only adopt ground-truth as positive information to guide the training of networks while ignoring negative information. Therefore, this paper proposes a network model to classify point cloud signals based on feature correlation and negative constraint, DANC-Net (dual-attention and negative constraint on point cloud classification). In the DANC-Net, the dual-attention mechanism is utilized to strengthen the interaction between local features of point cloud signal from both channel and space, thereby improving the expression ability of extracted features. Moreover, during the training of the DANC-Net, the negative constraint loss function ensures that the features in the same categories are close and those in the different categories are far away from each other in the representation space, so as to improve the feature extraction capability of the network. Experiments demonstrate that the DANC-Net achieves better classification performance than the existing point cloud classification algorithms on synthetic datasets ModelNet10 and ModelNet40 and real-scene dataset ScanObjectNN. The code is released at https://github.com/sunhang1986/DANC-Net.