Siyu Zhang

张思宇

Algorithm Engineer
Image Derivative Inc.
Email(org): zhangsiyu[at]idr[dot]ai
Research Interests
  • 3D Vision
  • Scene Understanding
  • Augmented Reality
About me

I'm currently working as an Algorithm Engineer on topics related to Neural Rendering at Image Derivative Inc.(像衍科技) where I'm one of the founding members. I'm also affiliated with State Key Lab of CAD & CG, Zhejiang University(ZJU), working closely with the 3D Vision Group led by Prof. Xiaowei Zhou.

I received my bachelor's degree in 2020 from the department of CSE, SUSTech. I spent three amazing years at 3DV&MR Group of SenseTime Research as Intern and Full-time Research Engineer, where I developed 3D Object Recognition algorithms for applications related Augmented Reality as well as participated in various research projects.

Research Interests

My interest lies in general 3D Computer Vision and Deep Learning with a special interest in understanding and digitalization of 3D scenes. Although working mostly as a full-stack research engineer to develop working solutions out of state-of-the-art technologies, I'm also fascinated with tackling fundamental challenges in 3D Vision through research and have engaged in a couple of exciting research projects.

Rescent News

    [Jul, 2020] Graduated from SUSTech and will be joining ZJU-Sensetime Joint Lab of 3D Vision as a full-time researcher.
    [Jan, 2020] One paper about 3D object detection accepted by CVPR 2020.
Publications

Disp

Stereo 3D Object Detection via Shape Prior GuidedInstance Disparity Estimation
Jiaming Sun*, Linghao Chen*, Yiming Xie, Siyu Zhang, Qinhong Jiang, Xiaowei Zhou. (*-indicates equal contributions)
CVPR 2020, TPAMI 2021
[PDF][Code]

Talk & Prensetations

Paper Review on Object 6D Pose Estimations:

  • [08/2020] Andvances in Object 6D Pose (mainly covered papers published in CVPR 20' and ECCV 20')[PDF]
  • [10/2020] Object 6D Pose Estimation by Action & Decision[PDF]
  • [04/2020] Object 6D Pose by Render & Compare[PDF]

Other topics:

  • [04/2021] Radiance-based Neural RGB-D reconstruction[PDF]