Sina Davari

I am a PhD candidate at the University of Toronto, working at the intersection of computer vision, deep learning, and generative AI for automated construction site safety.

My research addresses the data scarcity problem in construction safety monitoring: real-world annotated datasets are expensive, dangerous to collect, and statistically rare for the hazards that matter most. I use diffusion models and ControlNet to synthesize photorealistic, fully-annotated construction site images — enabling robust worker detection, PPE recognition, and hazard identification without costly on-site data collection.

Outside engineering, I come from a six-generation lineage of poets and literature scholars. I sing Persian classical vocal (Avaz) and play the Setar, and I bring that same love of Persian arts and culture to applying generative AI for the analysis of classical poetry and digital humanities.