Gen3DEval: Using vLLMs for Automatic Evaluation of Generated 3D Objects
Fuente:
arXiv
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866913788593176576 |
|---|---|
| author | Maiti, Shalini Agapito, Lourdes Kokkinos, Filippos |
| author_facet | Maiti, Shalini Agapito, Lourdes Kokkinos, Filippos |
| contents | Rapid advancements in text-to-3D generation require robust and scalable evaluation metrics that align closely with human judgment, a need unmet by current metrics such as PSNR and CLIP, which require ground-truth data or focus only on prompt fidelity. To address this, we introduce Gen3DEval, a novel evaluation framework that leverages vision large language models (vLLMs) specifically fine-tuned for 3D object quality assessment. Gen3DEval evaluates text fidelity, appearance, and surface quality by analyzing 3D surface normals, without requiring ground-truth comparisons, bridging the gap between automated metrics and user preferences. Compared to state-of-the-art task-agnostic models, Gen3DEval demonstrates superior performance in user-aligned evaluations, placing it as a comprehensive and accessible benchmark for future research on text-to-3D generation. The project page can be found here: \href{https://shalini-maiti.github.io/gen3deval.github.io/}{https://shalini-maiti.github.io/gen3deval.github.io/}. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_08125 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Gen3DEval: Using vLLMs for Automatic Evaluation of Generated 3D Objects Maiti, Shalini Agapito, Lourdes Kokkinos, Filippos Computer Vision and Pattern Recognition Rapid advancements in text-to-3D generation require robust and scalable evaluation metrics that align closely with human judgment, a need unmet by current metrics such as PSNR and CLIP, which require ground-truth data or focus only on prompt fidelity. To address this, we introduce Gen3DEval, a novel evaluation framework that leverages vision large language models (vLLMs) specifically fine-tuned for 3D object quality assessment. Gen3DEval evaluates text fidelity, appearance, and surface quality by analyzing 3D surface normals, without requiring ground-truth comparisons, bridging the gap between automated metrics and user preferences. Compared to state-of-the-art task-agnostic models, Gen3DEval demonstrates superior performance in user-aligned evaluations, placing it as a comprehensive and accessible benchmark for future research on text-to-3D generation. The project page can be found here: \href{https://shalini-maiti.github.io/gen3deval.github.io/}{https://shalini-maiti.github.io/gen3deval.github.io/}. |
| title | Gen3DEval: Using vLLMs for Automatic Evaluation of Generated 3D Objects |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2504.08125 |