3CAD: A Large-Scale Real-World 3C Product Dataset for Unsupervised Anomaly
Enquan Yang, Peng Xing, Hanyang Sun, Wenbo Guo, Yuanwei Ma, Zechao Li, Dan Zeng
2025-02-14
Summary
This paper talks about 3CAD, a new dataset for detecting defects in 3C products (computers, communication devices, and consumer electronics) using AI. It's much larger and more realistic than previous datasets, with over 27,000 high-resolution images of real manufacturing defects.
What's the problem?
Current datasets for teaching AI to spot defects in products are too small and don't represent real-world manufacturing problems well. They don't have enough examples of different types of defects or show how defects actually look in real factories. This makes it hard for researchers to create AI that can accurately detect problems in real manufacturing settings.
What's the solution?
The researchers created 3CAD, a huge dataset with images from actual 3C production lines. It includes eight different types of product parts and shows a wide variety of defects, including multiple defects in a single image. They also developed a new AI method called CFRG that uses this dataset to detect defects more accurately, especially small ones that are hard to spot.
Why it matters?
This matters because better defect detection can improve the quality of electronics we use every day. By providing a more realistic and challenging dataset, 3CAD helps researchers develop AI that can work in real factories, not just in labs. This could lead to fewer defective products reaching consumers and help companies save money by catching problems earlier in the manufacturing process.
Abstract
Industrial anomaly detection achieves progress thanks to datasets such as MVTec-AD and VisA. However, they suf- fer from limitations in terms of the number of defect sam- ples, types of defects, and availability of real-world scenes. These constraints inhibit researchers from further exploring the performance of industrial detection with higher accuracy. To this end, we propose a new large-scale anomaly detection dataset called 3CAD, which is derived from real 3C produc- tion lines. Specifically, the proposed 3CAD includes eight different types of manufactured parts, totaling 27,039 high- resolution images labeled with pixel-level anomalies. The key features of 3CAD are that it covers anomalous regions of different sizes, multiple anomaly types, and the possibility of multiple anomalous regions and multiple anomaly types per anomaly image. This is the largest and first anomaly de- tection dataset dedicated to 3C product quality control for community exploration and development. Meanwhile, we in- troduce a simple yet effective framework for unsupervised anomaly detection: a Coarse-to-Fine detection paradigm with Recovery Guidance (CFRG). To detect small defect anoma- lies, the proposed CFRG utilizes a coarse-to-fine detection paradigm. Specifically, we utilize a heterogeneous distilla- tion model for coarse localization and then fine localiza- tion through a segmentation model. In addition, to better capture normal patterns, we introduce recovery features as guidance. Finally, we report the results of our CFRG frame- work and popular anomaly detection methods on the 3CAD dataset, demonstrating strong competitiveness and providing a highly challenging benchmark to promote the development of the anomaly detection field. Data and code are available: https://github.com/EnquanYang2022/3CAD.