arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2406.18541cs.CVcs.GR
收藏

通过样本选择改进3D点云法向量估计

Refining 3D Point Cloud Normal Estimation via Sample Selection

School of Information Science and Technology, Dalian Maritime University(大连海事大学信息科学技术学院);Institute of Artificial Intelligence, Beijing University of Technology(北京工业大学人工智能研究院);School of Science, Zhejiang Sci-Tech University(浙江理工大学理学院);School of Mathematical Sciences, Dalian University of Technology(大连理工大学数学科学学院)

Jun Zhou, Yaoshun Li, Hongchen Tan, Mingjie Wang, Nannan Li, Xiuping Liu

AI总结:

本文提出一种基于样本选择的3D点云法向量估计框架,通过引入全局信息、约束机制和基于置信度的样本选择策略提升模型鲁棒性,并在定向和非定向任务中达到最先进性能。

摘要:

近年来,点云法向量估计作为一种经典且基础的算法,在3D几何处理领域引起了广泛关注。尽管当前基于神经网络的方法取得了显著性能,但其鲁棒性仍受训练数据质量和模型性能的影响。在本研究中,我们设计了一个用于法向量估计的基础框架,通过引入全局信息和多种约束机制来增强现有模型。此外,我们采用基于置信度的策略来选择合理样本,以进行公平且鲁棒的网络训练。引入的样本置信度可整合到损失函数中,以平衡不同样本对模型训练的影响。最后,我们利用现有的定向方法来校正估计的非定向法向量,在定向和非定向任务中均达到了最先进的性能。大量实验结果表明,我们的方法在广泛使用的基准上表现良好。

英文摘要:

In recent years, point cloud normal estimation, as a classical and foundational algorithm, has garnered extensive attention in the field of 3D geometric processing. Despite the remarkable performance achieved by current Neural Network-based methods, their robustness is still influenced by the quality of training data and the models' performance. In this study, we designed a fundamental framework for normal estimation, enhancing existing model through the incorporation of global information and various constraint mechanisms. Additionally, we employed a confidence-based strategy to select the reasonable samples for fair and robust network training. The introduced sample confidence can be integrated into the loss function to balance the influence of different samples on model training. Finally, we utilized existing orientation methods to correct estimated non-oriented normals, achieving state-of-the-art performance in both oriented and non-oriented tasks. Extensive experimental results demonstrate that our method works well on the widely used benchmarks.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑