Tianyuan Huang

Email: tianyuah [at] stanford [dot] edu

tianyuan_prof.jpg

Hi, I’m Tianyuan Huang (黄天元), a Software Engineer at Waymo on the Perception Scene Understanding team, contributing to the Waymo onboard driver. I received my Ph.D from Stanford University, co-advised by Ram Rajagopal (Electrical Engineering/Civil Engineering) and Jackelyn Hwang (Sociology). Before coming to Stanford, I obtained my B.Eng. degree from South China University of Technology studying Urban Planning and Computer Science.

My research uses multimodal learning to extract societal signals from large-scale spatial-temporal data: street-level imagery, satellite observations, and textual sources that capture how built environments change over time. The goal is to make visible what was previously unmeasurable at scale, and turn those observations into findings that matter for the people and policies they touch.

Publications

  1. Built environment disparities are amplified during extreme weather recovery
    Tianyuan Huang, Chad Zanocco, Zhecheng Wang, Jackelyn Hwang, and Ram Rajagopal
    Nature
  2. SkyScript: A Large and Semantically Diverse Vision-Language Dataset for Remote Sensing
    Zhecheng Wang, Rajanie Prabha*, Tianyuan Huang*, Jiajun Wu, and Ram Rajagopal
    AAAI 2024
  3. CityPulse: Fine-Grained Assessment of Urban Change with Street View Time Series
    Tianyuan Huang, Zejia Wu, Jiajun Wu, Jackelyn Hwang, and Ram Rajagopal
    AAAI 2024 Special Track on AI for Social Impact
  4. A System for Automated Vehicle Damage Localization and Severity Estimation Using Deep Learning
    Yuntao Ma, Hiva Ghanbari, Tianyuan Huang, Jeremy Irvin, Oliver Brady, Sofian Zalouk, Hao Sheng, Andrew Ng, Ram Rajagopal, and Mayur Narsude
    IEEE Transactions on Intelligent Transportation Systems
  5. Estimating building energy efficiency from street view imagery, aerial imagery, and land surface temperature data
    Kevin Mayer, Lukas Haas, Tianyuan Huang, Juan Bernabé-Moreno, Ram Rajagopal, and Martin Fischer
    Applied Energy
  6. Detecting Neighborhood Gentrification at Scale via Street-level Visual Data
    Tianyuan Huang, Timothy Dai, Zhecheng Wang, Hesu Yoon, Hao Sheng, Andrew Y. Ng, Ram Rajagopal, and Jackelyn Hwang
    IEEE Big Data 2022
  7. Detecting Neighborhood Gentrification at Scale via Street Views and POIs (Student Abstract)
    Tianyuan Huang
    AAAI 2022
  8. SCHMEAR: Scalable Construction of Holistic Models for Energy Analysis from Rooftops
    Thomas R. Dougherty, Tianyuan Huang, Yirong Chen, Rishee K. Jain, and Ram Rajagopal
    ACM BuildSys 2021
  9. Learning Neighborhood Representation from Multi-Modal Multi-Graph: Image, Text, Mobility Graph and Beyond
    Tianyuan Huang, Zhecheng Wang, Hao Sheng, Andrew Y. Ng, and Ram Rajagopal
    KDD 2021 Workshop on Deep Learning for Spatiotemporal Data, Applications, and Systems (DeepSpatial)
    Best Paper Runner-Up
  10. SRC: Discovering Human Activity Community in A City
    Tianyuan Huang
    ACM SIGSPATIAL 2019