Synthesizing Reality: Leveraging the Generative AI-Powered Platform Midjourney for Construction Worker Detection
This addresses data scarcity for construction safety monitoring, but it is incremental as it applies an existing generative AI tool to a new domain.
This study tackled the problem of inadequate data diversity and volume for construction worker detection by generating 12,000 synthetic images using Midjourney, resulting in a model that achieved average precisions of 0.937 and 0.642 on real construction images at IoU thresholds of 0.5 and 0.5 to 0.95, respectively.
While recent advancements in deep neural networks (DNNs) have substantially enhanced visual AI's capabilities, the challenge of inadequate data diversity and volume remains, particularly in construction domain. This study presents a novel image synthesis methodology tailored for construction worker detection, leveraging the generative-AI platform Midjourney. The approach entails generating a collection of 12,000 synthetic images by formulating 3000 different prompts, with an emphasis on image realism and diversity. These images, after manual labeling, serve as a dataset for DNN training. Evaluation on a real construction image dataset yielded promising results, with the model attaining average precisions (APs) of 0.937 and 0.642 at intersection-over-union (IoU) thresholds of 0.5 and 0.5 to 0.95, respectively. Notably, the model demonstrated near-perfect performance on the synthetic dataset, achieving APs of 0.994 and 0.919 at the two mentioned thresholds. These findings reveal both the potential and weakness of generative AI in addressing DNN training data scarcity.