DMD-Net: Deep Mesh Denoising Network

2025-06-28

AI Summary

This paper addresses 3D mesh denoising via a novel end-to-end deep learning framework, DMD-Net. Methodologically, it introduces a dual-branch network operating on both primal and dual graphs of the mesh, leveraging graph convolutions to extract geometric features; a primal–dual fusion module enables cross-domain feature complementarity. Furthermore, a feature-guided Transformer is incorporated to adaptively model long-range geometric dependencies and produce robust intermediate representations. The core contributions lie in (i) a unified primal–dual graph joint modeling paradigm and (ii) a feature-driven Transformer architecture tailored for mesh geometry. Extensive experiments on large-scale 3D mesh datasets demonstrate that DMD-Net consistently achieves state-of-the-art or competitive performance across diverse noise levels—particularly excelling under high noise, where it preserves fine-grained geometric details and topological consistency more effectively than existing methods.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

We present Deep Mesh Denoising Network (DMD-Net), an end-to-end deep learning framework, for solving the mesh denoising problem. DMD-Net consists of a Graph Convolutional Neural Network in which aggregation is performed in both the primal as well as the dual graph. This is realized in the form of an asymmetric two-stream network, which contains a primal-dual fusion block that enables communication between the primal-stream and the dual-stream. We develop a Feature Guided Transformer (FGT) paradigm, which consists of a feature extractor, a transformer, and a denoiser. The feature extractor estimates the local features, that guide the transformer to compute a transformation, which is applied to the noisy input mesh to obtain a useful intermediate representation. This is further processed by the denoiser to obtain the denoised mesh. Our network is trained on a large scale dataset of 3D objects. We perform exhaustive ablation studies to demonstrate that each component in our network is essential for obtaining the best performance. We show that our method obtains competitive or better results when compared with the state-of-the-art mesh denoising algorithms. We demonstrate that our method is robust to various kinds of noise. We observe that even in the presence of extremely high noise, our method achieves excellent performance.

Problem

• Develops deep learning framework for mesh denoising
• Uses primal-dual fusion for improved denoising performance
• Robust to various noise types including extreme cases

Innovation

• Asymmetric two-stream network with primal-dual fusion
• Feature Guided Transformer for mesh transformation
• Graph Convolutional Neural Network for denoising