Zixiang Zhou

Zixiang Zhou

Software Engineer at Google

Zixiang Zhou (周子翔)

I am currently a Software Engineer at Google, working on foundational models. Prior to this, I earned my Ph.D. in Computer Science from the University of Central Florida, where I was advised by Dr. Hassan Foroosh. I also worked at Luminar Technologies, focusing on LiDAR perception models. My research interests include Machine Learning, 3D Computer Vision, LiDAR Perception, and Foundational Models.

News

[2025.01] Joined Google as a Software Engineer working on foundational models.
[2024.01] Two papers accepted in ICRA 2024.
[2023.12] I successfully defended my PhD thesis, titled “Towards a Robust and Efficient Deep Neural Network for the Lidar Point Cloud Perception”.
[2023.06] One paper accepted in IROS 2023.
[2023.06] LPFormer gets the top performance in the Waymo Open Dataset Pose Estimation challenge 2023. arXiv, Youtube.
[2022.12] LidarMultiNet is accepted as a oral paper in AAAI 2023.
[2022.09] The official implementation of CenterFormer is publicly available on GitHub.
[2022.09] Our extended paper of LidarMultiNet is released on arXiv. LidarMultiNet currently has the SOTA performances on 5 major LiDAR perception benchmarks.

  • Waymo LiDAR semantic segmentation (71.1 mIOU)
  • Waymo 3D detection (79.9 mAPH)
  • nuScenes 3D detection (71.6 NDS)
  • nuScenes LiDAR segmentation (81.4 mIoU)
  • nuScenes panoptic segmentation (81.4 PQ)

[2022.07] One oral paper accepted in ECCV 2022.
[2022.05] LidarMultiNet is the official winner of the Waymo Open Dataset 3D semantic segmentation challenge 2022

Work Experience

  • Jan. 2025 - Present: Google
    • Software Engineer
  • Sep. 2023 - Jan. 2025: Luminar Technologies
    • Senior Machine Learning Engineer
  • May. 2021 - Dec. 2022: Tusimple
    • Deep Perception Research Intern