PeopleofSuperintelligence

Researcher

Daeho Kim

Daeho Kim is a civil engineering researcher who applies computer vision and AI to human–robot collaboration on construction sites. He studied architectural engineering at Yonsei and civil engineering at Michigan before joining the University of Toronto as an assistant professor in 2021. His work includes camera-based proximity monitoring, synthetic construction imagery, and Pixel2Graph, a neural network for detecting contact hazards between workers and robots. As of October 2026, he directs the Construction Vision Lab and teaches construction students about AI, robotics and digital twinning.

Timeline

  1. 2006–2016

    From Seoul to Michigan

    He comes from Seoul, South Korea, and begins his academic path at Yonsei University in 2006. He earns bachelor's and master's degrees in architectural engineering there. In 2016, he begins a graduate path in civil engineering at the University of Michigan that includes MSc and PhD studies.

  2. 2019–2020

    Proximity monitoring for safety

    His 2019 paper describes remote proximity monitoring between mobile construction resources using camera-mounted UAVs. In 2020, he coauthors a paper on predicting mobile-object proximity to prevent contact-driven accidents in co-robotic construction.

  3. 2021

    Assistant professor at Toronto

    In July 2021, he joins the University of Toronto's Department of Civil & Mineral Engineering as an assistant professor of civil engineering (Construction), focused on co-robotic construction. During his PhD, he develops worker-safety technologies using imagery, computer vision and deep neural networks.

  4. 2022

    A robot for digital twins

    In 2022, a University of Toronto funding announcement includes his project to prototype a fully autonomous mobile robot for construction digital twinning with higher-order DNNs. His team proposes synthetic virtual and miniature-scale construction images to train visual AI for construction sites.

  5. 2024

    Pixel2Graph detects hazards

    In 2024, he coauthors a hazard-detection study that uses the two-in-one DNN architecture Pixel2Graph to detect construction objects and their relationships. Tests on real-site images report 90.63% recall@5 when object bounding boxes and classes are given.

  6. As of October 2026

    Assistant professor and lab director

    As of October 2026, he is an assistant professor in U of T's Department of Civil & Mineral Engineering and lab director of the Construction Vision Lab. His research aims to improve safe collaboration between construction workers and robots through scalable safety technologies, redesigned workspaces and training. In August 2026, he teaches in a construction-management program where courses give students hands-on experience with digital twinning, BIM and robotics while they explore AI in construction.

Career

  1. Since 2021University of TorontoAssistant Professor
  2. NowUniversity of TorontoDirector, Construction Vision Lab

Education

  1. From 2016University of MichiganMSc in civil engineering
  2. From 2006Yonsei UniversityBachelor's in architectural engineering
  3. Yonsei UniversityMaster's in architectural engineering
  4. University of MichiganPhD in civil engineering

From the biography.

Checked career sources: civmin.utoronto.ca.

Colleagues

People with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Daeho Kim yet. Send us a free-licensed photo.

  1. Meet Daeho Kim, one of CivMin's new faculty members - Department of Civil & Mineral Engineering civmin.utoronto.ca. Checked 27 October 2026.
  2. Daeho Kim - Department of Civil & Mineral Engineering civmin.utoronto.ca. Checked 27 October 2026.
  3. ‘Sputtertron’, among latest CFI-funded U of T projects, could help develop new materials for a greener economy | University of Toronto utoronto.ca. Checked 27 October 2026.
  4. U of T prof seeks higher-order construction robotics using AI canada.constructconnect.com. Checked 27 October 2026.
  5. Single-Shot Visual Relationship Detection for the Accurate Identification of Contact-Driven Hazards in Sustainable Digitized Construction - Rutgers, The State University of New Jersey researchwithrutgers.org. Checked 27 October 2026.
  6. Team – Construction Vision Lab cvl.civmin.utoronto.ca. Checked 27 October 2026.
  7. U of T students explore AI’s role in construction safety, productivity through master’s program canada.constructconnect.com. Checked 27 October 2026.