Industrial Object Detection Leveraging Synthetic Data for Training Deep Learning Models

The increasing adoption of synthetic training data has emerged as a promising solution in various domains, owing to its ability to provide accurately labeled datasets at a lower cost compared to manually annotated real-world data. In this study, we explore the utilization of synthetic data for training deep learning models in the field of industrial object detection. Our objective is to evaluate the performance of different models trained with varying ratios of real and synthetic data, aiming to identify the optimal ratio that yields superior results. Additionally, we investigate the impact of introducing randomization to the synthetic data on the overall performance of the trained models. The findings from our research contribute to understand the role of synthetic data in industrial object detection.