MeshCNN: a network with an edge
By Rana Hanocka, Amir Hertz, Noa Fish, Raja Giryes, Shachar Fleishman, Daniel Cohen-Or. ACM Transactions on Graphics, 2019
375 citations recorded by the source corpus.
Polygonal meshes provide an efficient representation for 3D shapes. They explicitly captureboth shape surface and topology, and leverage non-uniformity to represent large flat regions as well as sharp, intricate features. This non-uniformity and irregularity, however, inhibits mesh analysis efforts using neural networks that combine convolution and pooling operations. In this paper, we utilize the unique properties of the mesh for a direct analysis of 3D shapes using MeshCNN, a convolutional neural network designed specifically for triangular meshes. Analogous to classic CNNs, MeshCNN combines specialized convolution and pooling layers that operate on the mesh edges, by leveraging their intrinsic geodesic connections. Convolutions are applied on edges and the four edges of their incident triangles, and pooling is applied via an edge collapse operation that retains surface topology, thereby, generating new mesh connectivity for the subsequent convolutions. MeshCNN learns which edges to collapse, thus forming a task-driven process where the network exposes and expands the important features while discarding the redundant ones. We demonstrate the effectiveness of MeshCNN on various learning tasks applied to 3D meshes.
Subfields: Computational Mechanics, Computer Graphics and Computer-Aided Design
View Paper: https://www.semanticscholar.org/paper/d47959e1f541d09f31dca8317b2f13c844379c4a
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
- Keenan Crane
- Souhaib Attaiki
- Nicholas Sharp
- M. Ovsjanikov
- Andrea Tagliasacchi
- Zhiqin Chen
- Geoffrey E. Hinton
- G. Peyr'e
In Computer Graphics and Computer-Aided Design
Citations in the corpus
- BSP-Net: Generating Compact Meshes via Binary Space Partitioning (2019, background): “One of the main reasons is that the nonuniformity and irregularity of triangle tessellations do not naturally support conventional convolution and pooling operations [19].”
- Learning Gradient Fields for Shape Generation (2020, neutral): “Common representations emerged for deep generative 3D modeling include voxel-based [23,63], mesh-based [3,48,20,27,35,57], and assembly-based techniques [34,41].”
- DiffusionNet: Discretization Agnostic Learning on Surfaces (2020, methodology): “…of graph-like approaches which specifically leverage mesh connectivity [Bod-nar et al. 2021; Feng et al. 2019; Gong et al. 2019; Hajij et al. 2020; Hanocka et al. 2019; Lim et al. 2018; Milano et al. 2020; Verma et al. 2018], the structure of discrete operators [Smirnov and Solomon 2021], or even…”