Don't Mesh with Me: Generating Constructive Solid Geometry Instead of Meshes by Fine-Tuning a Code-Generation LLM
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866912361815736320 |
|---|---|
| author | Mews, Maximilian Aynetdinov, Ansar Schiller, Vivian Eisert, Peter Akbik, Alan |
| author_facet | Mews, Maximilian Aynetdinov, Ansar Schiller, Vivian Eisert, Peter Akbik, Alan |
| contents | While recent advancements in machine learning, such as LLMs, are revolutionizing software development and creative industries, they have had minimal impact on engineers designing mechanical parts, which remains largely a manual process. Existing approaches to generating 3D geometry most commonly use meshes as a 3D representation. While meshes are suitable for assets in video games or animations, they lack sufficient precision and adaptability for mechanical engineering purposes. This paper introduces a novel approach for the generation of 3D geometry that generates surface-based Constructive Solid Geometry (CSG) by leveraging a code-generation LLM. First, we create a dataset of 3D mechanical parts represented as code scripts by converting Boundary Representation geometry (BREP) into CSG-based Python scripts. Second, we create annotations in natural language using GPT-4. The resulting dataset is used to fine-tune a code-generation LLM. The fine-tuned LLM can complete geometries based on positional input and natural language in a plausible way, demonstrating geometric understanding. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_15279 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Don't Mesh with Me: Generating Constructive Solid Geometry Instead of Meshes by Fine-Tuning a Code-Generation LLM Mews, Maximilian Aynetdinov, Ansar Schiller, Vivian Eisert, Peter Akbik, Alan Machine Learning Graphics While recent advancements in machine learning, such as LLMs, are revolutionizing software development and creative industries, they have had minimal impact on engineers designing mechanical parts, which remains largely a manual process. Existing approaches to generating 3D geometry most commonly use meshes as a 3D representation. While meshes are suitable for assets in video games or animations, they lack sufficient precision and adaptability for mechanical engineering purposes. This paper introduces a novel approach for the generation of 3D geometry that generates surface-based Constructive Solid Geometry (CSG) by leveraging a code-generation LLM. First, we create a dataset of 3D mechanical parts represented as code scripts by converting Boundary Representation geometry (BREP) into CSG-based Python scripts. Second, we create annotations in natural language using GPT-4. The resulting dataset is used to fine-tune a code-generation LLM. The fine-tuned LLM can complete geometries based on positional input and natural language in a plausible way, demonstrating geometric understanding. |
| title | Don't Mesh with Me: Generating Constructive Solid Geometry Instead of Meshes by Fine-Tuning a Code-Generation LLM |
| topic | Machine Learning Graphics |
| url | https://arxiv.org/abs/2411.15279 |