CAD 100K: A Comprehensive Multi-Task Dataset for Car Related Visual Anomaly Detection

2026-04-10

AI Summary

This study addresses the absence of a unified multitask benchmark for visual anomaly detection in automotive manufacturing, which hinders comprehensive model evaluation and generalization. To bridge this gap, the authors introduce the first large-scale multitask dataset tailored for automotive anomaly detection, encompassing seven vehicle subdomains and three core tasks, augmented with synthetic data to facilitate few-shot learning. Systematic experiments demonstrate that multitask learning effectively promotes cross-task knowledge transfer, while also revealing potential task conflicts. This work not only establishes a standardized evaluation platform but also lays the foundation for multitask modeling in industrial visual anomaly detection.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Multi-task visual anomaly detection is critical for car-related manufacturing quality assessment. However, existing methods remain task-specific, hindered by the absence of a unified benchmark for multi-task evaluation. To fill in this gap, We present the CAD Dataset, a large-scale and comprehensive benchmark designed for car-related multi-task visual anomaly detection. The dataset contains over 100 images crossing 7 vehicle domains and 3 tasks, providing models a comprehensive view for car-related anomaly detection. It is the first car-related anomaly dataset specialized for multi-task learning(MTL), while combining synthesis data augmentation for few-shot anomaly images. We implement a multi-task baseline and conduct extensive empirical studies. Results show MTL promotes task interaction and knowledge transfer, while also exposing challenging conflicts between tasks. The CAD dataset serves as a standardized platform to drive future advances in car-related multi-task visual anomaly detection.

Problem

• multi-task learning
• visual anomaly detection
• car manufacturing
• benchmark dataset
• quality assessment

Innovation

• multi-task learning
• visual anomaly detection
• synthetic data augmentation
• car manufacturing quality
• benchmark dataset