期刊
ACM TRANSACTIONS ON GRAPHICS
卷 41, 期 3, 页码 -出版社
ASSOC COMPUTING MACHINERY
DOI: 10.1145/3507905
关键词
Geometric deep learning; geometry processing; discrete differential geometry; partial differential equations
资金
- Fields Institute for Research in Mathematical Sciences
- Vector Institute for AI
- NSF
- ERC [758800]
- ANR AI Chair AIGRETTE
- Packard Fellowship
- NSF CAREER Award [1943123]
- Direct For Computer & Info Scie & Enginr
- Div Of Information & Intelligent Systems [1943123] Funding Source: National Science Foundation
We introduce a new general-purpose approach to deep learning on three-dimensional surfaces, which is based on the insight that a simple diffusion layer is highly effective for spatial communication. The resulting networks are simple, robust, and efficient, and can automatically adapt to changes in resolution and sampling of a surface.
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.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据