Photo

Cheng Lin 林铖

Researcher, Tencent

chlin@connect.hku.hk

I'm currently a researcher at Tencent Game AI Research Center. I received my Ph.D. from The Univeristy of Hong Kong (HKU), advised by Prof. Wenping Wang. Before that, I did my bachelor's degree in Shandong University. My research interests include geometric modeling, 3D vision, shape analysis and computer graphics.

[Full CV]   [Github]

Publications


Adaptive Surface Normal Constraint for Depth Estimation
ICCV 2021
Point2Skeleton: Learning Skeletal Representations from Point Clouds
CVPR 2021 (Oral)
Learnable Motion Coherence for Correspondence Pruning
CVPR 2021
Modeling 3D Shapes by Reinforcement Learning
ECCV 2020
Most discussed CS papers on Twitter [Hot Computer Science]
SEG-MAT: 3D Shape Segmentation Using Medial Axis Transform
IEEE TVCG 2020
Adaptive Compact Attention For Few-shot Video-to-video Translation
Arxiv 2020
Floorplan-Jigsaw: Jointly Estimating Scene Layout and Aligning Partial Scans
ICCV 2019
Image-based Reconstruction of Wire Art
SIGGRAPH 2017 (TOG)
Supporting easy physical-to-virtual creation of mobile vr maze games: a new genre
CHI 2017
A Novel Workbench for Collaboratively Constructing 3D Virtual Environment
IIKI 2017

Awards


  • Computer Science Postgraduate Scholarship [2016, 2018]
  • Postgraduate Scholarship [2016-2020]
  • Google Excellence Scholarship [2015]
  • President’s Scholarship (the highest honor in Shandong University) [2015]
  • Baosteel Scholarship [2015]
  • CCF (China Computer Federation) Top 100 Outstanding CS Students of China [2015]
  • National Scholarship of China [2013, 2014, 2015]
  • Experiences


  • Visual Computing Group, Technical University of Munich (TUM)
    Research Visit
    Advisor: Prof. Matthias Nießner
    May 2019 - Oct 2019

  • Research Center of Digital Media Technology, Ministry of Education
    Research Assistant
    Advisor: Prof. Chenglei Yang
    Dec 2014 - June 2016
  • Invited Talks


  • Point2Skeleton: Learning Skeletal Representations of Point Clouds
    VALSE Webinar
    August, 2021

  • State-of-the-art Applications of Medial Axis Transform in 3D Vision and Shape Analysis
    Advanced Computer Graphics, ShanghaiTech University
    March, 2021

  • Modeling 3D Shapes by Reinforcement Learning
    Holistic Scene Structures for 3D Vision, ECCV 2020 Workshop
    August, 2020
  • Patents


  • A mixed-reality game system based on ubiquitous computing
    China Invention Patent (Granted) CN105373224B

  • A collaborative scene modeling system based on wide-angle folding touch screens
    China Invention Patent (Granted) CN106406430B

  • A collaborative scene modeling system based on mobile terminals and holographic display
    China Invention Patent (Granted) CN104794752B





  • ©Cheng Lin.  Last updated: Sep. 19, 2021.