4 files



FireNet is an open ML training dataset for visual recognition of fire safety equipment. Our dataset directly links the objects to their respective Uniclass, the construction sector’s classification scheme to name objects.

FireNet has been designed as a ML training dataset for experimentation and therefore fulfills multiple machine learning scenarios (classification, object detection, semantic segmentation). It is intended as a domain specific dataset to refine pre-trained standard architectures.

Uniclass & Description

Pr_40_50_28_64 Portable fire extinguishers

Pr_75_75_30_50 Manual call points

Pr_40_50_28_29 Fire protective blankets

Pr_40_10_77_32 Fire escape route signs

Pr_40_10_77_31 Fire equipment signs

Pr_75_75_30_97 Visual alarm signal devices Pr_75_75_30_30 Fire alarm sounders

Pr_75_75_30_65 Point smoke detectors


Centre for Digital Built Britain (CDBB), under InnovateUK grant number RG96233