The XFMP dataset for Explainable Fine-Grained Abnormal Behavior Recognition on Medical Personal Protective Equipment

The XFMP dataset is designed for abnormal behavior recognition on medical personal protective equipment, which features fine-grained and explainable annotations. The proper use of medical personal protective equipment (MPPE) is critical for frontline healthcare workers (HCWs) to handle highly contagious diseases. Due to the complexity of PPE donning and doffing protocols, public health organizations typically recommend having trained observers monitor the entire PPE donning and doffing process, preventing self-contamination and transmission. However, the high costs of manual monitoring impede the implementation of this practice, which makes AI-assisted PPE monitoring highly valuable. To address this, we propose an explainable and fine-grained dataset for MPPE doffing monitoring called the XFMP dataset. The dataset contains 3596 expert-annotated samples over three sub-tasks: doffing stage classification (DSC), abnormal action recognition (AAR), and critical region localization (CRL).