Fig 1.
Overall network structure.
Fig 2.
Feature Enhancement Module.
Fig 3.
Prior Processing Module.
Fig 4.
Discriminator architecture based on self-attention.
Fig 5.
Scale factor applied prior to softmax operation.
Fig 6.
(a) Neighborhood selection for point p. (b) PCA-based normal computation (red arrows). (c) Local coordinate system construction.
Fig 7.
Comparison of the effect of generating chair-like point clouds using different methods.
Table 6.
Ablation Study on Keypoint Sampling Ratio for Curvature Estimation.
Table 1.
Experimental settings.
Table 2.
Quantitative comparison of point cloud data generated for chairs and airplanes.
Table 3.
Comparison of training, reasoning speed and model parameter quantity.
Fig 8.
Objective function variation curve of chair category point cloud generation process.
Fig 9.
The generation process of point clouds for chair categories.
Fig 10.
Generating the point cloud of aircraft categories.
Table 4.
Classification and verification test.
Table 5.
Comparative Analysis of Normal Estimation Methods.
Table 7.
Generate ablation experiment results for two categories chairs and airplanes.
Table 8.
Quantitative comparison of other methods for modifying discriminators.