Point cloud denoising method based on iterativePFN

Yaxu Zhang; Li Cui

doi:10.1117/12.3055717

9 January 2025 Point cloud denoising method based on iterativePFN

Yaxu Zhang, Li Cui

Author Affiliations +

Proceedings Volume 13486, Fourth International Conference on Computer Vision, Application, and Algorithm (CVAA 2024); 134860Y (2025) https://doi.org/10.1117/12.3055717
Event: Fourth International Conference on Computer Vision, Application, and Algorithm (CVAA 2024), 2024, Chengdu, China

Abstract

3D point clouds acquired by using advanced scanning equipment have been used in a variety of applications like automated vehicles, 3D modeling, and cultural heritage preservation. Due to limitations of the devices, point cloud data is often contaminated with noise. This paper focuses on denoising point clouds for downstream tasks. We adopt a self-attention mechanism to enhance feature extraction in the encoding phase. A layer normalization method is proposed to address the issue of gradient explosion occurring during the iterative process of point clouds denoising. Experimental results demonstrate that our method can diminish the error values of CD and P2M compared with the previous methods. Visualization results demonstrate that our method notably enhances the smoothness of the denoised point clouds.

(2025) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Yaxu Zhang and Li Cui "Point cloud denoising method based on iterativePFN", Proc. SPIE 13486, Fourth International Conference on Computer Vision, Application, and Algorithm (CVAA 2024), 134860Y (9 January 2025); https://doi.org/10.1117/12.3055717

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

;

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE