AttPNet: Attention-Based Deep Neural Network for 3D Point Set Analysis

Abstract

Point set is a major type of 3D structure representation format characterized by its data availability and compactness. Most former deep learning-based point set models pay equal attention to different point set regions and channels, thus having limited ability in focusing on small regions and specific channels that are important for characterizing the object of interest. In this paper, we introduce a novel model named Attention-based Point Network (AttPNet). It uses attention mechanism for both global feature masking and channel weighting to focus on characteristic regions and channels. There are two branches in our model. The first branch calculates an attention mask for every point. The second branch uses convolution layers to abstract global features from point sets, where channel attention block is adapted to focus on important channels. Evaluations on the ModelNet40 benchmark dataset show that our model outperforms the existing best model in classification tasks by 0.7% without voting. In addition, experiments on augmented data demonstrate that our model is robust to rotational perturbations and missing points. We also design a Electron Cryo-Tomography (ECT) point cloud dataset and further demonstrate our model’s ability in dealing with fine-grained structures on the ECT dataset.

Hierarchical Point Cloud Encoding and Decoding With Lightweight Self-Attention Based Model

“…Convolution-based methods have also been applied to point cloud data [16], [17] by directly performing convolution-like operation on 3D points. More recently, graph-based neural networks [18], [19] and attentionbased methods [20], [21], [22] also demonstrate good performance for point cloud feature learning. However, these related works only learn features in encoding manner.…”

“…PointNet [8] 1k p 3.5 89.2 PointNet++ [9] 1k p 1.5 90.7 PointNet++ [9] 5k p,n 1.5 91.9 Kd-Net [35] 32k p 2.0 91.8 PointCNN [16] 1k p 0.3 92.5 DGCNN [18] 1k p 1.8 92.9 PCNN [36] 1k p 1.4 92.3 DensePoint [33] 1k p 0.7 93.2 RS-CNN [37] 1k p 1.4 93.6 KPConv [17] 7k p 14.3 92.9 Pnt Tfmer [24] 1k p,n 13.50 92.8 Pnt Tfmer [20] 1k p 9.6 93.7 PCT [21] 1k p 2.9 93.2 PointASNL [25] 1k p -92.9 AttPNet [22] 1k p -93.6 PAT [23] 1k p -91.7…”

Game theory and MCDM-based unsupervised sentiment analysis of restaurant reviews

“…Unsupervised approaches for sentiment analysis include hierarchical and partition methods. Due to the high computation cost, hierarchical algorithms are not ideal for massive datasets [74][75][76][77][78][79]. Partition methods are relatively scalable and straightforward but have poor accuracy and stability and are sensitive to noisy data.…”

DNet: Dynamic Neighborhood Feature Learning in Point Cloud

“…For example, if the first point has 10 neighborhood points with the shape of (1,10,3), while the second point has 20 neighborhood points with the shape of (1,20,3), the network cannot stack these two points for learning. However, if both of the shapes of the two points is (1,20,3), the network can stack the two points into the shape of (2,20,3). Therefore, in this paper, the number of initial neighborhood points is fixed to k, and the mask M i j is used to remove the pseudo neighborhood points from the neighborhood since these points are not conducive to the network learning of the local region.…”

“…With the rapid development of three dimensional (3D) sensing technologies, using deep learning to understand and analyze point clouds is becoming one of the important research topics [1][2][3]. As the output of 3D sensor, point cloud is composed of much number of points in 3D space.…”