Avatar

Jingwei Tang (唐经纬)

Ph.D. Student

Computer Graphics Laboratory
    ETH Zürich
    CAB G 86.2
    Universitätsstrasse 6
    CH-8092 Zürich
    Switzerland

jingwei.tang(at)inf.ethz.ch

+41 (0) 44 632 07 56


Short Bio

In 2019, I joined the Computer Graphics Laboratory (CGL) as a Ph.D. student with Professor Markus Gross at ETH Zürich. In 2018, I received my M.Sc. degree in Computational Science and Engineering at at ETH Zürich. My B.Sc. degree in Physics was received in 2016 at Nanjing University.

My research interests lie in computer graphics and machine learning, especially using data-driven methods to solve physically-based fluid simulation problems.


Educations

Ph.D. / ETH Zürich
Computer Science
2019 - Present
M.Sc. / ETH Zürich
Computational Science and Engineering
2016 - 2018
B.Sc. / Nanjing University
Physics
2012 - 2016


News

Our learning-based sampling for natural matting paper on CVPR 2019 is now online.
Check out the project webpage.
11 Apr 2019

I joined Computer Graphics Laboratory at ETH Zürich as a Ph.D. student.
I will be working on machine learning methods for physically-based fluid simulations.

01 Jan 2019

I finished my Master Thesis at Disney Research Zürich .
The thesis is on Data-driven methods for Image Matting. My supervisors are Yağız Aksoy, Dr. Tunç Aydın , Dr. Cengiz Öztireli and Prof. Markus Gross.

30 Aug 2018

Research Topics

Fluid Simulation and Machine Learning

Our research focues on combining data-driven methods from Machine Learning and physically-based simulation, especially fluid simulation.


Data-driven Image Matting

The goal of natural image matting is the estimation of accurate opacities of a user-defined foreground object. We tackle this problem with data-driven methods. In particular, we developed a CNN-based matting pipeline that consists of sampling network to predict foreground and background colors and a matting network to predict alpha mattes.

Avatar Avatar Avatar