DiffusionNet: Discretization Agnostic Learning on Surfaces

By Nicholas Sharp, Souhaib Attaiki, Keenan Crane, M. Ovsjanikov. ACM Transactions on Graphics, 2020

312 citations recorded by the source corpus.

We introduce a new general-purpose approach to deep learning on three-dimensional surfaces based on the insight that a simple diffusion layer is highly effective for spatial communication. The resulting networks are automatically robust to changes in resolution and sampling of a surface—a basic property that is crucial for practical applications. Our networks can be discretized on various geometric representations, such as triangle meshes or point clouds, and can even be trained on one representation and then applied to another. We optimize the spatial support of diffusion as a continuous network parameter ranging from purely local to totally global, removing the burden of manually choosing neighborhood sizes. The only other ingredients in the method are a multi-layer perceptron applied independently at each point and spatial gradient features to support directional filters. The resulting networks are simple, robust, and efficient. Here, we focus primarily on triangle mesh surfaces and demonstrate state-of-the-art results for a variety of tasks, including surface classification, segmentation, and non-rigid correspondence.

Subfields: Computational Mechanics, Computer Graphics and Computer-Aided Design

View Paper: https://www.semanticscholar.org/paper/5ed7420fc4280860098f414d3b0a666b518e79e9

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

1. Matthias Nießner
2. Angela Dai
3. Alexey Artemov
4. J. Susskind
5. Charles Ruizhongtai Qi
6. Leonidas J. Guibas
7. L. Zampieri
8. Andreas Geiger

In Computer Graphics and Computer-Aided Design

1. Angela Dai
2. Matthias Nießner
3. Navdeep Jaitly
4. Andreas Geiger
5. Yaron Lipman

Citations in the corpus

• Generating Molecular Conformer Fields (2023, methodology): “In order to do this, we compute the k leading eigenvectors of the normalized graph Laplacian ∆ G (Maskey et al., 2022; Sharp et al., 2022) as positional encoding for points in the graph.”
• MeshGPT: Generating Triangle Meshes with Decoder-Only Transformers (2023, background): “While several discriminative approaches capable of learning signals directly on mesh structure were proposed over the recent years [16, 21, 23, 33, 40, 52, 56], direct mesh generation remains underex-plored.”