# Fully Convolutional Mesh Autoencoder using Efficient Spatially Varying Kernels

Source: [Fully Convolutional Mesh Autoencoder using Efficient Spatially Varying Kernels](<https://www.sota2.com/research/paper/68fa68881e209131cea1cb55>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T22:39:59.706Z

Authors: Yi Zhou, Chenglei Wu, Zimo Li, Chen Cao, Yuting Ye, Jason Saragih, Hao Li, Yaser Sheikh

Published: 2020-06-08T02:30:13.000Z

## Abstract

Learning latent representations of registered meshes is useful for many 3D tasks. Techniques have recently shifted to neural mesh autoencoders. Although they demonstrate higher precision than traditional methods, they remain unable to capture fine-grained deformations. Furthermore, these methods can only be applied to a template-specific surface mesh, and is not applicable to more general meshes, like tetrahedrons and non-manifold meshes. While more general graph convolution methods can be employed, they lack performance in reconstruction precision and require higher memory usage. In this paper, we propose a non-template-specific fully convolutional mesh autoencoder for arbitrary registered mesh data. It is enabled by our novel convolution and (un)pooling operators learned with globally shared weights and locally varying coefficients which can efficiently capture the spatially varying contents presented by irregular mesh connections. Our model outperforms state-of-the-art methods on reconstruction accuracy. In addition, the latent codes of our network are fully localized thanks to the fully convolutional structure, and thus have much higher interpolation capability than many traditional 3D mesh generation models.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Unconditional mesh shape generation on Animal SMAL](<https://www.sota2.com/research/sota/unconditional-mesh-shape-generation-on-animal-smal.md>): Unconditional mesh shape generation; Animal SMAL; Error Rate (ER): 8.01.

- [Unconditional Mesh Shape Generation on Human](<https://www.sota2.com/research/sota/unconditional-mesh-shape-generation-on-human.md>): Unconditional mesh shape generation; Human; ER: 5.43.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2006.04325>)

[PDF](<https://arxiv.org/pdf/2006.04325.pdf>)