Chi Wang 王驰

Chi Wang 王驰

Ph.D. Candidate

Zhejiang University

I am a student of Computer Science and Technology at the CAD&CG State Key Lab, Zhejiang University, supervised by Prof. Weiwei Xu. My primary research interests lie in semantic segmentation, image matting, image generation, 3D generation ,and novel view synthesis.

Interests
  • Artificial Intelligence
  • Machine Learning
  • Computer Vision
  • Computer Graphics
Education
  • Ph.D. in Computer Science and Technology, 2019

    Zhejiang University

  • MEng in Computer Science and Technology, 2017

    Beijing Institute of Technology

  • BSc in Software Engineering, 2013

    Beijing Institute of Technology

Publications

Weakly Supervised Image Matting via Patch Clustering
ICIG 2023 (Oral) [Oral presentation 10.27% acceptance rate; overall 30.31%]

Active Boundary Loss for Semantic Segmentation
AAAI 2022 (Oral) [Oral presentation 5.5% acceptance rate; overall 15.0%]

Attention-guided Temporally Coherent Video Object Matting
ACM MM 2021 [27.9% acceptance rate]

Automatic liver segmentation using multi-plane integrated fully convolutional neural networks
BIBM 2018 [19.3% acceptance rate]

Industry Experience

 
 
 
 
 
Kuaishou Technology
Research Intern
June 2023 – Present Hangzhou

Responsibilities include:

  • Exploring AIGC, mainly 3D generation.
 
 
 
 
 
Alimama [Alibaba Group]
Research Intern
October 2021 – Present Beijing

Responsibilities include:

  • Exploring the challenge of generating high-quality commercial fonts; Costs reduction and efficiency improvement for non-standardized font development
  • Participating in and completing two projects
  • Released 2 fonts:阿里妈妈东方大楷阿里妈妈刀隶体
  • A technical summary paper accepted by CVPR 2023, and one national invention patent
 
 
 
 
 
DAMO Academy [Alibaba Group]
Research Intern
September 2019 – October 2021 Hangzhou

Responsibilities include:

  • Explore high-quality algorithms for matting and semantic segmentation
  • Achieving a first-place ranking in the well-known large-scale semantic segmentation dataset ADE20K using the company account
  • Two technical summary papers: one accepted for oral presentation at AAAI 2022 and another accepted by MM 2021, as well as one national invention patent