Saved in:
Bibliographic Details
Main Authors: Bregeon, Germain, Preda, Marius, Ispas, Radu, Zaharia, Titus
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.02125
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910038125182976
author Bregeon, Germain
Preda, Marius
Ispas, Radu
Zaharia, Titus
author_facet Bregeon, Germain
Preda, Marius
Ispas, Radu
Zaharia, Titus
contents In this paper, we introduce a novel 3D mesh convolution-based autoencoder for geometry compression, able to deal with irregular mesh data without requiring neither preprocessing nor manifold/watertightness conditions. The proposed approach extracts meaningful latent representations by learning features directly from the mesh faces, while preserving connectivity through dedicated pooling and unpooling operations. The encoder compresses the input mesh into a compact base mesh space, which ensures that the latent space remains comparable. The decoder reconstructs the original connectivity and restores the compressed geometry to its full resolution. Extensive experiments on multi-class datasets demonstrate that our method outperforms state-of-the-art approaches in both 3D mesh geometry reconstruction and latent space classification tasks. Code available at: github.com/germainGB/MeshConv3D
format Preprint
id arxiv_https___arxiv_org_abs_2603_02125
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle A 3D mesh convolution-based autoencoder for geometry compression
Bregeon, Germain
Preda, Marius
Ispas, Radu
Zaharia, Titus
Computer Vision and Pattern Recognition
In this paper, we introduce a novel 3D mesh convolution-based autoencoder for geometry compression, able to deal with irregular mesh data without requiring neither preprocessing nor manifold/watertightness conditions. The proposed approach extracts meaningful latent representations by learning features directly from the mesh faces, while preserving connectivity through dedicated pooling and unpooling operations. The encoder compresses the input mesh into a compact base mesh space, which ensures that the latent space remains comparable. The decoder reconstructs the original connectivity and restores the compressed geometry to its full resolution. Extensive experiments on multi-class datasets demonstrate that our method outperforms state-of-the-art approaches in both 3D mesh geometry reconstruction and latent space classification tasks. Code available at: github.com/germainGB/MeshConv3D
title A 3D mesh convolution-based autoencoder for geometry compression
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2603.02125