profile.jpg

Connor Lee

I’m a final-year PhD candidate in the Autonomous Robotics and Controls Laboratory at the California Institute of Technology (Caltech) where I develop computer vision algorithms for robot localization and semantic perception.

I’ve also spent time at Google AR working on motion tracking algorithms for augmented reality devices and at Apple where I worked on Siri visual intelligence.

Before pursuing my Ph.D., I worked as a software engineer at Microsoft on the Identity team. I earned my B.S. in Computer Science from Caltech.

News

Mar 23, 2024 New paper describing our fast and free method for generating semantic segmentation annotations for aerial imagery using satellite land cover data and visual foundation models.
Mar 19, 2024 Released a new RGB-thermal dataset, CART, which is the first RGB-T dataset created for aerial field robotic perception algorithms in natural environments.
Nov 28, 2023 Our paper RGB-X Object Detection via Scene-Specific Fusion Modules has been accepted to WACV 2024.

Selected publications

  1. autoseg.jpg
    Semantics from Space: Satellite-Guided Thermal Semantic Segmentation Annotation for Aerial Field Robots
    Connor Lee ,  Saraswati Soedarmadji ,  Matthew Anderson ,  Anthony J. Clark , and 1 more author
    arXiv preprint arXiv:2403.14056, Mar 2024
  2. cart.jpg
    CART: Caltech Aerial RGB-Thermal Dataset in the Wild
    Connor Lee* ,  Matthew Anderson* ,  Nikhil Raganathan ,  Xingxing Zuo , and 3 more authors
    arXiv preprint arXiv:2403.08997, Mar 2024
  3. dsf.jpg
    RGB-X Object Detection via Scene-Specific Fusion Modules
    Sri Aditya Deevi* ,  Connor Lee* ,  Lu Gan* ,  Sushruth Nagesh , and 2 more authors
    In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision , Jan 2024
  4. river.jpg
    Online Self-Supervised Thermal Water Segmentation for Aerial Vehicles
    Connor Lee ,  Jonathan Gustafsson Frennert ,  Lu Gan ,  Matthew Anderson , and 1 more author
    In IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) , Oct 2023

    IROS Best Paper Award on Agri-Robotics (Finalist)

  5. thermal-uda.jpg
    Unsupervised RGB-to-Thermal Domain Adaptation via Multi-Domain Attention Network
    Lu Gan ,  Connor Lee ,  and  Soon-Jo Chung
    In IEEE International Conference on Robotics and Automation (ICRA) , Jun 2023
  6. landmark.jpg
    Self-Supervised Landmark Discovery for Terrain-Relative Navigation
    Connor Lee ,  Esmir Mesic ,  and  Soon-Jo Chung
    In ICRA Workshop on Unconventional spatial representations: Opportunities for robotics , Jun 2023
  7. sr-short.jpg
    A seasonally invariant deep transform for visual terrain-relative navigation
    Anthony Fragoso ,  Connor Lee ,  Austin McCoy ,  and  Soon-Jo Chung
    Science Robotics, Jun 2021