PU-EdgeFormer: Edge Transformer for Dense Prediction in Point Cloud Upsampling

TL;DR

PU-EdgeFormer introduces a novel combination of graph convolution and transformer architecture to improve dense point cloud upsampling, effectively capturing local and global structures for superior performance.

Contribution

It proposes a new EdgeFormer unit combining graph convolution and multi-head self-attention for enhanced point cloud upsampling.

Findings

Outperforms existing state-of-the-art methods in upsampling quality.

Effectively captures local geometry and global structure.

Demonstrates superior results in subjective and objective evaluations.

Abstract

Despite the recent development of deep learning-based point cloud upsampling, most MLP-based point cloud upsampling methods have limitations in that it is difficult to train the local and global structure of the point cloud at the same time. To solve this problem, we present a combined graph convolution and transformer for point cloud upsampling, denoted by PU-EdgeFormer. The proposed method constructs EdgeFormer unit that consists of graph convolution and multi-head self-attention modules. We employ graph convolution using EdgeConv, which learns the local geometry and global structure of point cloud better than existing point-to-feature method. Through in-depth experiments, we confirmed that the proposed method has better point cloud upsampling performance than the existing state-of-the-art method in both subjective and objective aspects. The code is available at https://github.com/dohoon2045/PU-EdgeFormer.