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

2025-03-18

AI Summary

Engineering-domain 3D generative models face deployment challenges due to low-quality training data, poor controllability, and high annotation costs. Method: This paper introduces the first high-fidelity, vehicle-engineering-specific 3D mesh dataset. We propose an automated quality-aware filtering pipeline integrating DINOv2 and SigLIP multimodal embeddings, caption semantic analysis, and uncertainty estimation—enabled by a dedicated quality classifier—to substantially improve in-domain data fidelity and consistency. Subsequently, we fine-tune the SV3D model on this curated dataset. Results: Experiments demonstrate significant improvements over conventional caption- or aesthetics-based filtering: generated meshes exhibit higher geometric accuracy, enhanced part-level controllability, and superior domain adaptation. Our approach establishes a reproducible, scalable foundation—both data and methodology—for engineering-grade 3D content generation.

Technology Category

• Computer Vision: 3D Computer Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Natural Language Processing: Generation

Application Category

• Web Mining and Content Analysis: Web data quality in the era of algorithmically-generated content
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Search and Retrieval-Augmented AI: Vertical and domain-specific search

Abstract

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.

Problem

• Limited accuracy and controllability of 3D generative models in engineering.
• Lack of high-quality domain-specific 3D datasets for fine-tuning.
• Bottleneck in data filtering and annotation for 3D generative modeling.

Innovation

• Automated data filtering using quality classifier
• Incorporates DINOv2 and SigLIP embeddings
• Refines data through caption-based analysis