# Point cloud generation adversarial network based on self-attention and curvature

## TL;DR

This paper introduces SAC-GAN, a new adversarial network for generating high-quality 3D point clouds using self-attention and curvature learning.

## Contribution

The novel SAC-GAN model combines self-attention and curvature mechanisms to improve point cloud generation quality and consistency.

## Key findings

- SAC-GAN outperforms existing models like TreeGAN and SP-GAN in terms of generation quality.
- The model achieves a 4.24 reduction in JSD and a 0.8 decrease in MMD, indicating improved performance.
- Generated point clouds show better shape integrity and authenticity compared to other methods.

## Abstract

As a mainstream form of 3D data, point clouds are widely used in computer vision for tasks such as segmentation, classification, and target detection due to their simple representation method and high stability and accuracy. Considering the issues of noise points and uneven point distribution in current generation models, we propose a novel adversarial network model, SAC-GAN, which incorporates both a self-attention mechanism and a curvature learning mechanism. Firstly, the feature enhancement module and the pre-processing module, which are based on the ShapeNetCore open dataset, are designed on the generator to enhance the authenticity of local geometric details in the generated point cloud. Secondly, the loss function of the discriminator is adjusted to combine the traditional Wasserstein distance with the normal vector of key points to guide the generation of subtle features of point clouds and improve the quality and consistency of generated point clouds; Finally, to enhance the discriminator’s capacity to extract both local and global features, a self-attention mechanism is introduced. This enhances the discriminator’s capacity to discern the details of the generated point cloud and offers superior feedback to the generator. Experimental results indicate that the proposed point cloud generation model outperforms existing methods, including TreeGAN, SP-GAN, PDGN, and WarpingGAN, in terms of generation quality. Specifically, the model achieves a reduction in JSD by 4.24, a decrease in MMD by 0.8, and an increase in COV by 1.25%. It can be proven that the point cloud generated by SAC-GAN model has a good performance in shape integrity and authenticity.