Vision language model for interpretable and fine-grained detection of safety compliance in diverse workplaces

Open Webpage

Zhiling Chen, Hanning Chen, Mohsen Imani, Ruimin Chen, Farhad Imani

Published: 2025, Last Modified: 16 Oct 2025Expert Syst. Appl. 2025EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Highlights•Introduce Clip2Safety for improved safety compliance in diverse workplaces.•Integrate scene recognition, prompts, and fine-grained verification in detection models.•Enhance real-time safety monitoring through a robust and adaptable compliance detection framework.•Achieve improved accuracy and speed in detecting PPE compliance across six real-world scenarios.