arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2605.00632cs.CVcs.AIcs.GRcs.HCcs.LG
收藏

BlenderRAG: 通过检索增强的代码合成实现高保真的3D物体生成

BlenderRAG: High-Fidelity 3D Object Generation via Retrieval-Augmented Code Synthesis

University of Bologna(博洛尼亚大学);Ferrari S.p.A.(法拉利公司)

Massimo Rondelli, Francesco Pivi, Maurizio Gabbrielli

AI总结:

BlenderRAG通过检索增强的代码合成方法,利用500个专家验证的多模态数据集提升Blender代码生成的准确性和一致性,无需微调或专用硬件。

摘要:

从自然语言自动生成可执行的Blender代码仍然具有挑战性,最先进的LLMs常产生语法错误和几何不一致的对象。我们提出了BlenderRAG,一个基于检索增强生成的系统,使用500个专家验证的多模态数据集(文本、代码、图像)覆盖50个物体类别。在生成过程中通过检索语义相似的例子,BlenderRAG将四个最先进的LLMs的编译成功率从40.8%提升到70.0%,语义归一化对齐从0.41提升到0.77(CLIP相似度),无需微调或专用硬件,使其能够立即部署。数据集和代码将在https://github.com/MaxRondelli/BlenderRAG上提供。

英文摘要:

Automatic generation of executable Blender code from natural language remains challenging, with state-of-the-art LLMs producing frequent syntactic errors and geometrically inconsistent objects. We present BlenderRAG, a retrieval-augmented generation system that operates on a curated multimodal dataset of 500 expert-validated examples (text, code, image) across 50 object categories. By retrieving semantically similar examples during generation, BlenderRAG improves compilation success rates from 40.8% to 70.0% and semantic normalized alignment from 0.41 to 0.77 (CLIP similarity) across four state-of-the-art LLMs, without requiring fine-tuning or specialized hardware, making it immediately accessible for deployment. The dataset and code will be available at https://github.com/MaxRondelli/BlenderRAG.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑