<br>SPU-Net：通过自投影优化从粗到精重建的自监督点云上采样,IEEE Transactions on Image Processing

当前位置： X-MOL 学术 › IEEE Trans. Image Process. › 论文详情

Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)

SPU-Net：通过自投影优化从粗到精重建的自监督点云上采样
IEEE Transactions on Image Processing ( IF 10.8 ) Pub Date : 6-13-2022 , DOI: 10.1109/tip.2022.3182266
Xinhai Liu ₁ , Xinchen Liu ₁ , Yu-Shen Liu ₁ , Zhizhong Han ₂

Affiliation

点云上采样的任务旨在从稀疏且不规则的点集中获取密集且均匀的点集。尽管深度学习模型取得了重大进展，但最先进的方法需要地面实况密集点集作为监督，这使得它们仅限于在合成配对训练数据下进行训练，而不适合在真实情况下进行训练。扫描稀疏数据。然而，从真实扫描的稀疏数据中获取大量成对的稀疏-密集点集作为监督是昂贵且乏味的。为了解决这个问题，我们提出了一种名为 SPU-Net 的自监督点云上采样网络，以捕获位于底层对象表面上的点的固有上采样模式。具体来说，我们提出了一个从粗到细的重建框架，它包含两个主要部分：分别是点特征提取和点特征扩展。在点特征提取中，我们将自注意力模块与图卷积网络（GCN）结合起来，以同时捕获局部区域内部和局部区域之间的上下文信息。在点特征扩展中，我们引入了一种分层可学习的折叠策略，以生成具有可学习的二维网格的上采样点集。此外，为了进一步优化生成的点集中的噪声点，我们提出了一种与均匀项和重建项相关的新颖的自投影优化作为联合损失，以促进自监督点云上采样。我们对合成数据集和真实扫描数据集进行了各种实验，结果表明我们实现了与最先进的监督方法相当的性能。

"点击查看英文标题和摘要"

SPU-Net: Self-Supervised Point Cloud Upsampling by Coarse-to-Fine Reconstruction With Self-Projection Optimization

The task of point cloud upsampling aims to acquire dense and uniform point sets from sparse and irregular point sets. Although significant progress has been made with deep learning models, state-of-the-art methods require ground-truth dense point sets as the supervision, which makes them limited to be trained under synthetic paired training data and not suitable to be under real-scanned sparse data. However, it is expensive and tedious to obtain large numbers of paired sparse-dense point sets as supervision from real-scanned sparse data. To address this problem, we propose a self-supervised point cloud upsampling network, named SPU-Net, to capture the inherent upsampling patterns of points lying on the underlying object surface. Specifically, we propose a coarse-to-fine reconstruction framework, which contains two main components: point feature extraction and point feature expansion, respectively. In the point feature extraction, we integrate the self-attention module with the graph convolution network (GCN) to capture context information inside and among local regions simultaneously. In the point feature expansion, we introduce a hierarchically learnable folding strategy to generate upsampled point sets with learnable 2D grids. Moreover, to further optimize the noisy points in the generated point sets, we propose a novel self-projection optimization associated with uniform and reconstruction terms as a joint loss to facilitate the self-supervised point cloud upsampling. We conduct various experiments on both synthetic and real-scanned datasets, and the results demonstrate that we achieve comparable performances to state-of-the-art supervised methods.

更新日期：2024-08-26

点击分享查看原文

点击收藏

公开下载

阅读更多本刊新发论文本刊介绍/投稿指南