Fully-Synthetic Training for Visual Quality Inspection in Automotive Production
This addresses the challenge of data scarcity and labeling costs for manufacturers, but is incremental as it builds on existing synthetic data methods.
The paper tackles the problem of costly and time-consuming data collection for visual quality inspection in manufacturing by proposing a pipeline for generating synthetic images using domain randomization, and demonstrates that an object detection model trained solely on this synthetic data outperforms models trained on real images in three real inspection scenarios.
Visual Quality Inspection plays a crucial role in modern manufacturing environments as it ensures customer safety and satisfaction. The introduction of Computer Vision (CV) has revolutionized visual quality inspection by improving the accuracy and efficiency of defect detection. However, traditional CV models heavily rely on extensive datasets for training, which can be costly, time-consuming, and error-prone. To overcome these challenges, synthetic images have emerged as a promising alternative. They offer a cost-effective solution with automatically generated labels. In this paper, we propose a pipeline for generating synthetic images using domain randomization. We evaluate our approach in three real inspection scenarios and demonstrate that an object detection model trained solely on synthetic data can outperform models trained on real images.