MeshFleet：面向特定领域生成建模的过滤与标注三维车辆数据集

MeshFleet: Filtered and Annotated 3D Vehicle Dataset for Domain Specific Generative Modeling

March 18, 2025

作者: Damian Boborzi, Phillip Mueller, Jonas Emrich, Dominik Schmid, Sebastian Mueller, Lars Mikelsons

cs.AI

摘要

生成模型在3D物體領域近期取得了顯著進展。然而，由於無法滿足特定領域任務所需的精確度、品質與可控性，這些模型在工程等領域的實際應用仍受限。對大型生成模型進行微調，是使其在這些領域中可用的前景方向。建立高品質、特定領域的3D數據集對於微調大型生成模型至關重要，但數據篩選與註釋過程仍是主要瓶頸。我們提出了MeshFleet，這是一個從Objaverse-XL（目前最廣泛的公開3D物體集合）中提取並經過篩選與註釋的3D車輛數據集。我們的方法基於品質分類器，提出了一套自動化數據篩選流程。該分類器在Objaverse的手動標註子集上訓練，結合了DINOv2與SigLIP嵌入，並通過基於標題的分析與不確定性估計進行了優化。我們通過與基於標題和圖像美學評分的技術進行對比分析，以及使用SV3D進行的微調實驗，展示了我們篩選方法的有效性，強調了針對特定領域的3D生成建模進行精準數據選擇的重要性。

English

Generative models have recently made remarkable progress in the field of 3D objects. However, their practical application in fields like engineering remains limited since they fail to deliver the accuracy, quality, and controllability needed for domain-specific tasks. Fine-tuning large generative models is a promising perspective for making these models available in these fields. Creating high-quality, domain-specific 3D datasets is crucial for fine-tuning large generative models, yet the data filtering and annotation process remains a significant bottleneck. We present MeshFleet, a filtered and annotated 3D vehicle dataset extracted from Objaverse-XL, the most extensive publicly available collection of 3D objects. Our approach proposes a pipeline for automated data filtering based on a quality classifier. This classifier is trained on a manually labeled subset of Objaverse, incorporating DINOv2 and SigLIP embeddings, refined through caption-based analysis and uncertainty estimation. We demonstrate the efficacy of our filtering method through a comparative analysis against caption and image aesthetic score-based techniques and fine-tuning experiments with SV3D, highlighting the importance of targeted data selection for domain-specific 3D generative modeling.