AIAM Software

MeshGPT software creates better AI-driven 3D models

The result of a study by researchers from the Technical University of Munich (TUM) and Politecnico di Torino, in collaboration with Audi AG

Stay up to date with everything that is happening in the wonderful world of AM via our LinkedIn community.

Researchers from the Technical University of Munich (TUM) and Politecnico di Torino, in collaboration with Audi AG, developed MeshGPT, a 3D mesh generation software that uses generative AI. Their innovative approach utilizes decoder-only transformers to create triangle meshes that are cleaner and more coherent, boasting a remarkable 9% increase in shape coverage and a 30-point boost in FID scores (a metric used to assess the quality of images created by a generative model), compared to existing methods. This means sharper edges, higher fidelity, and compact designs that set a new quality standard. The full paper is available here.

MeshGPT creates triangle meshes by autoregressively sampling from a transformer model that has been trained to produce tokens from a learned geometric vocabulary. These tokens can then be decoded into the faces of a triangle mesh. MeshGPT reflects the compactness typical of artist-created meshes, in contrast to dense triangle meshes extracted by iso-surfacing methods from neural fields. Inspired by recent advances in powerful large language models, the researchers adopted a sequence-based approach to autoregressively generate triangle meshes as sequences of triangles.

Researchers from the Technical University of Munich (TUM) and Politecnico di Torino, in collaboration with Audi AG developed MeshGPT

The first step was to learn a vocabulary of latent quantized embeddings, using graph convolutions, which inform these embeddings of the local mesh geometry and topology. The term “quantized embeddings” refers to a technique used in natural language processing (NLP) and machine learning to represent and store word embeddings or continuous-valued vectors in a quantized, discrete form.

These embeddings are sequenced and decoded into triangles by a decoder, ensuring that they can effectively reconstruct the mesh. A transformer is then trained on this learned vocabulary to predict the index of the next embedding given previous embeddings. Once trained, this model can be autoregressively sampled to generate new triangle meshes, directly generating compact meshes with sharp edges, more closely imitating the efficient triangulation patterns of human-crafted meshes.

Researchers from the Technical University of Munich (TUM) and Politecnico di Torino, in collaboration with Audi AG developed MeshGPT

MeshGPT demonstrates a notable improvement over state-of-the-art mesh generation methods, with a 9% increase in shape coverage and a 30-point enhancement in FID scores across various categories. Existing approaches often either miss details produce over-triangulated meshes, or output too simplistic shapes. The MeshGPT method can also be used to generate 3D assets for scenes.

Study authors include: Yawar Siddiqui, Antonio Alliegro, Alexey Artemov, Tatiana Tommasi, Daniele Sirigatti, Vladislav Rosov, Angela Dai, Matthias Nießner

Research
Composites AM 2024

746 composites AM companies individually surveyed and studied. Core composites AM market generated over $785 million in 2023. Market expected to grow to $7.8 billion by 2033 at 25.8% CAGR. This new...

Davide Sher

Since 2002, Davide has built up extensive experience as a technology journalist, market analyst and consultant for the additive manufacturing industry. Born in Milan, Italy, he spent 12 years in the United States, where he completed his studies at SUNY USB. As a journalist covering the tech and videogame industry for over 10 years, he began covering the AM industry in 2013, first as an international journalist and subsequently as a market analyst, focusing on the additive manufacturing industry and relative vertical markets. In 2016 he co-founded London-based VoxelMatters. Today the company publishes the leading news and insights websites VoxelMatters.com and Replicatore.it, as well as VoxelMatters Directory, the largest global directory of companies in the additive manufacturing industry.

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button
Close Popup
Privacy Settings saved!
Privacy Settings

When you visit any web site, it may store or retrieve information on your browser, mostly in the form of cookies. Control your personal Cookie Services here.

These cookies are necessary for the website to function and cannot be switched off in our systems.

Technical Cookies
In order to use this website we use the following technically required cookies
  • PHPSESSID
  • wordpress_test_cookie
  • wordpress_logged_in_
  • wordpress_sec

Decline all Services
Save
Accept all Services

Newsletter

Join our 12,000+ Professional community and get weekly AM industry insights straight to your inbox. Our editor-curated newsletter equips executives, engineers, and end-users with crucial updates, helping you stay ahead.