A Survey of Deep Learning-based Point Cloud Denoising

2025-08-23

AI Summary

Point cloud denoising faces challenges including diverse noise sources, degraded geometric fidelity, and compromised performance in downstream tasks. This paper presents a systematic survey of deep learning–based point cloud denoising methods. We propose the first unified functional taxonomy, categorizing approaches along two orthogonal dimensions: supervision level (supervised vs. unsupervised) and modeling paradigm (explicit, implicit, or learning-based). To enable rigorous comparison, we establish a standardized evaluation benchmark with consistent training configurations, assessing methods across four key dimensions: denoising quality, surface fidelity, point distribution uniformity, and computational efficiency. Comprehensive experiments on state-of-the-art methods reveal critical bottlenecks in current performance—particularly in preserving fine geometric structures under severe noise. Our analysis identifies three promising research directions: integration of geometric priors, self-supervised representation learning, and structure-aware optimization strategies.

Technology Category

• Machine Learning: Learning with Manifolds
• Computer Vision: 3D Computer Vision
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Accurate 3D geometry acquisition is essential for a wide range of applications, such as computer graphics, autonomous driving, robotics, and augmented reality. However, raw point clouds acquired in real-world environments are often corrupted with noise due to various factors such as sensor, lighting, material, environment etc, which reduces geometric fidelity and degrades downstream performance. Point cloud denoising is a fundamental problem, aiming to recover clean point sets while preserving underlying structures. Classical optimization-based methods, guided by hand-crafted filters or geometric priors, have been extensively studied but struggle to handle diverse and complex noise patterns. Recent deep learning approaches leverage neural network architectures to learn distinctive representations and demonstrate strong outcomes, particularly on complex and large-scale point clouds. Provided these significant advances, this survey provides a comprehensive and up-to-date review of deep learning-based point cloud denoising methods up to August 2025. We organize the literature from two perspectives: (1) supervision level (supervised vs. unsupervised), and (2) modeling perspective, proposing a functional taxonomy that unifies diverse approaches by their denoising principles. We further analyze architectural trends both structurally and chronologically, establish a unified benchmark with consistent training settings, and evaluate methods in terms of denoising quality, surface fidelity, point distribution, and computational efficiency. Finally, we discuss open challenges and outline directions for future research in this rapidly evolving field.

Problem

• Denoising raw point clouds corrupted by real-world noise
• Recovering clean point sets while preserving underlying structures
• Handling diverse and complex noise patterns in 3D data

Innovation

• Deep learning neural network architectures
• Unified benchmark with consistent training
• Functional taxonomy by denoising principles