Biography

My Chinese name is 李睿. I am currently a Ph.D. student in KAUST, Computational Imaging Group, working with Prof. Wolfgang Heidrich. Before that, I obtained my bachelor’s degree in Xidian University, School of Telecommunications Engineering, and then went to University of Science and Technology of China (USTC), School of Information Science and Technology, as a master’s student.

Research: my academic research interests are using various image measurements (e.g., light field, polarized images, normal RGB images) for higher-level challenging applications, such as photo-realistic 3D reconstruction/representation, viewpoint synthesis, depth estimation, reflection removal via hierarchical representation of neural networks in large-scale data setting.

Else: I also put much time into high-frequency trading and algorithmic trading by AI agent system for better automatic decision making and predicting the near-future actions of trading rivals, currently run a very small private fund in cryptocurrency markets. If you are a daily trader, you will actually meet my trading bots in most mainstream markets (I close them due to market regulation issues recently). Later, I am considering combining blockchain and 3D reconstruction tasks for a photo-realistic NFT as well.

Download my resumé.

Interests
  • Computational Photography
  • Photo-realistic 3D Reconstruction
  • Neural Rendering
Education
  • Ph.D. in Electrical and Computer Engineering, 2016-Present

    KAUST

  • M.S. in Information and Communication Engineering, 2013-2016

    University of Science and Technology of China

  • BSc in Communication Engineering, 2009-2013

    Xidian University

Experience

 
 
 
 
 
Founder
Mar 2021 – Present Internet
make digital art in blockchain, especiall from real scene data.
 
 
 
 
 
Founder
Oct 2017 – Present Internet
make digital art in blockchain, especiall from real scene data.
 
 
 
 
 
Sadeem
Research Intern
May 2017 – Oct 2017 Jeddah, Saudi Arabia
firmware developer, algorithm developer for vechicle detection by utral sonic sensor.
 
 
 
 
 
HKUST
Research Assistant
Feb 2016 – Aug 2016 Hong Kong
design vision tracking, object detection algorithms in a low-cost UAVs system, deploy first generation active tracking for Phatom 4 and Mavic.
 
 
 
 
 
DJI
Research Intern
Sep 2014 – Aug 2015 Shen Zhen, China
design vision tracking, object detection algorithms in a low-cost UAVs system, deploy first generation active tracking for Phatom 4 and Mavic.
 
 
 
 
 
State Key Laboratory of Integrated Services Networks
Research Assistant
State Key Laboratory of Integrated Services Networks
Mar 2013 – May 2013 Xi'an, China
Front-end engineer, design web page for visualizing vechicle, transportation data.

Talks

Skills

c

Skilled

c++

Skilled

Skilled

pandas

Skilled

cuda

Preliminary

scipy

Skilled

Contact

Search Publications

Quickly discover relevant content by filtering publications.

Machine Learning Logos

This repository hosts SVG vector versions of various ML library logos for convenience.

The logos are or may be trademarks of their respective owners.

Aside from the NumPy and Pytorch logos, the logos were hand-vectorized from raster sources with utmost love and caring.

Maintained by Valohai machine learning platform.

Caffe

Based on the typography of the Caffe site.

Font family: PT Serif Color: #C30000

Darknet

Based on the typography in the Darknet logo.

Font family: Luminari

Keras

Vectorized version of the Keras logo from Wikimedia.

Color: #D00000

Numpy

From the Numpy repository, with some fixes and minification applied.

Pytorch

From the Pytorch repository, with removal of CSS and minification applied.

TensorFlow

Based on the raster logo.

Font family: Interstate Regular

Theano

Based on theano_logo_allblue.svg with the text flattened to curves.

Font family: DejaVu Sans Mono

Torch

Based on torch-logo.png.

Font family: Titling Gothic FB Thin

MXNet

Based on mxnet2.png

Font Family: Titillium Web Regular Italic / Titillium Web Semibold Italic

CUDA

Based on NVIDIA_CUDA_V_2C_r.jpg

Matplotlib

Based on logo2.svg

Sympy

Based on sympy.svg

Scikit-learn

Based on scikit-learn-logo.svg

Caffe2

Based on an Illustrator image trace of the original .png logo.

PaddlePaddle

Original creator’s version.

DL4J

Reconstructed by hand from DL4J.png.

Color from logo.png.

OpenVINO Toolkit

openvino-toolkit.svg is based on the logotype, as on the site, set in Intel Clear Pro using Intel blue (#0071c5).

openvino-2.svg is based on this logo. The text is set in Montreal Medium, which is not exactly the same, but very close.

ImageNet

Based on the official logo.

Set in slightly fattened (using stroke width) Myriad Pro Regular.

The fattening caused minor artifacts around the corners of glyphs, but it doesn’t look too bad.

Pandas

Via https://github.com/pandas-dev/pandas/blob/761bceb77d44aa63b71dda43ca46e8fd4b9d7422/web/pandas/static/img/pandas.svg