Yan Xu

About Me

Hi! My name is Yan Xu. I am a software engineer at Waymo working on 3D object detection for autonomous driving scenarios. My job includes 3D detection model development, training, evaluation, and deployment onto autonomous driving vehicles. Before joining Waymo, I obtained my Ph.D. from Carnegie Mellon University where I was working on Computer Vision with \href{http://www.cs.cmu.edu/~kkitani/}{\underline{Prof. Kris Kitani}}. My research focus was multi-view multi-person 3D human pose estimation. Specifically, I have been working on using deep learning and geometry tools to develop 3D human pose estimation methods that does not require manual 3D training label annotation and can work beyond laboratory settings with calibrated camera networks. Thank you for your time and interests! Please find more about my work in the project and publication sections.

News

[Dec. 2022] I started my job at Waymo.
[Oct. 2022] Our paper on 3D human pose estimation with uncalibrated cameras is accepted to BMVC 2022.
[Jul. 2021] Our paper on cross-camera multi-target association is accepted to ICCV 2021.
[March. 2021] Our paper on widebaseline camera pose estimation is accepted to CVPR 2021.
[Feb. 2021] I'll be joining Waymo for Summer 2021 working with Zhenyao Zhu and Feiyu Chen.
[Aug. 2020] Our Few-Shot learning paper is released at Arxiv.
[Jan. 2020] Our Monocular camera pose estimation paper is accepted to WACV 2020.