The Story…


About Taichi

Previously I implemented ~40 graphics papers on physically based animation and rendering, almost from scratch in C++ on my five different code bases. However, gradually I find it important to have a unified software framework for physically based computer graphics, since simulation & rendering actually share a significant amount of similar code and unified software framework help both parts enhance each other - a tighter coupling here may actually open up some potential research opportunities.

Recently, I get some time so I decided to open-source, integrate and finally polish the code so that interested people can play with it.

I would like to thank Prof. Toshiya Hachisuka and Prof. Seiichi Koshizuka for making possible my internship at UTokyo, where I developed an initial part of Taichi; and to Prof. Hui Zhang, who supported me to submit further part of Taichi as the course project for Digital Media.

About Me


Yuanming Hu  胡渊鸣

(My last name is pronounced as "who" instead of "hue"; yes, I'll hopefully be Dr. "Who" in 5-7 years.)

Senior undergrad at Tsinghua University
Special Pilot Computer Science Class (Yao Class)
Expected Graduation: July 2017   GPA: 91/100 (5th / 33)

Research Interest: computer graphics
Email:                      LinkedIn

Programming Contests

Asia-Pacific Informatics Olympiad 2012, China District

China National Olympiad in Informatics 2012

ACM-ICPC Asia Changsha Regional Contest 2013, with my team

ACM-ICPC Asia Shanghai Regional Contest 2014, with my team

International Olympiad in Informatics 2013

China National Olympiad in Informatics 2014

Gold Medal, 1st / 450

Gold Medal, 13th / 350

Gold Medal, 3rd / 200

Gold Medal, 5th / 250

China Team Candidate

Problem Setter


Publicity department of the students’ union in Yao Class
As a member (2014.4-2015.3) and then the head (2015.3-2015.9), in charge of photography and graphic design.

Research Intern@Face++ 2015.3-2015.6
Face recognition research using deep learning.

Technical Assistant Intern@Oriental DreamWorks 2015.7-2015.8
Software engineering for render farm task monitoring and auto-translation plugins.


Graphics Engineer@Phantouch Technology 2015.9-2015.12
Built a real-time graphics system with low motion-to-photon latency.

Internship Student@the University of Tokyo 2016.3-2016.6
Computer graphics research (physically based animation and rendering) with Prof. Seiichi Koshizuka and Prof. Toshiya Hachisuka. I implemented 10+ graphics papers during this internship and designed a software framework for unified simulation and rendering, which turns out to be important components of Taichi.


Undergraduate Visiting Researcher@Stanford University 2016.6-2016.8
Computer vision research with Prof. Ron Fedkiw. We developed a novel line segment based reconstruction algorithm for thin tree twigs which outperforms state of the art.


Research Intern@Microsoft Research Asia 2016.9-
Computer vision research with Sr. Researcher Dr. Stephen Lin and Dr. Baoyuan Wang. We developed a novel color constancy algorithm that outperforms previous state of the art with a 100-fold speed increase. Results are accepted as an oral paper to CVPR 2017.



FC^4:Fully Convolutional Color Constancy with Confidence-weighted Pooling
Yuanming Hu, Baoyuan Wang, Stephen Lin

CVPR'17, Oral


Programming Languages:

        Proficient in C/C++(14), Python 2/3, GLSL, Java, CoffeeScript/JavaScript

        Capable of Mathematica, Haskell, CUDA, HTML/CSS, Ruby, Bash

Tools: OpenGL, Vim, Git, svn, Qt, ssh, TensorFlow, LaTeX, CMake, Visual Studio, CLion, tmux

(Natural) Languages: Chinese (native), English

        TOEFL: 112/120 (speaking 26)

        GRE: 327/340 (verbal 157, quantitive 170, AW 3.5)


Photography, graphic design, animated films, creating and destroying everything in my physically based simulators