Welcome!

I am Raymond Yeh, a Ph.D. candidate in Computational Imaging Group and Statistical Speech Technology Group of University of Illinois at Urbana-Champaign (UIUC), advised by Prof. Minh Do and Prof. Mark Hasegawa-Johnson.

I am interested in research relating to machine learning, deep learning, and signal processing, including audio, speech, and image processing.

I received my B.S. degree in Electrical Engineering from University of Illinois at Urbana-Champaign.

Contact Information
E-mail: yeh17 at illinois dot edu

Current and Past Affiliations

2010 - present

Summer 2016

Summer 2015

Summer 2014

Summer 2013




Preprints

Semantic Facial Expression Editing using Autoencoded Flow

Raymond Yeh, Ziwei Liu, Dan B Goldman, Aseem Agarwala

arXiv preprint, 2016


Semantic Image Inpainting with Perceptual and Contextual Losses

Raymond Yeh, Chen Chen, Teck Yian Lim, Mark Hasegawa-Johnson, Minh N. Do

arXiv preprint, 2016

Publications


Stable and Symmetric Filter Convolutional Neural Network

Raymond Yeh, Mark Hasegawa-Johnson, Minh N. Do

IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2016




Résumé

Education

University of Illinois at Urbana-Champaign

Doctor of Philosophy in Electrical Engineering, August 2014 - current

  • Advisors: Prof. Minh N. Do and Prof. Mark Hasegawa-Johnson

University of Illinois at Urbana-Champaign

Master of Science in Electrical Engineering, August 2014 - August 2016

  • Advisors: Prof. Minh N. Do and Prof. Mark Hasegawa-Johnson

University of Illinois at Urbana-Champaign

Bachelor of Science in Electrical Engineering, minor in Computer Science

  • Cumulative GPA: 4.00/4.00, August 2010 - May 2014
  • Graduated with Highest Honor
  • Advisor: Prof. Mark Hasegawa-Johnson

Research and Work Experience

Google

Software Engineering Intern, May 2016 - August 2016
  • Facial expression editing using deep learning.

Google

Software Engineering Intern, May 2015 - August 2015
  • Used Google Brain to predict user response on an in-ad mute survey.

Johns Hopkins Univ., Human Language Technology Center of Excellence

Research Intern, Summer Camp for Applied Language Exploration (SCALE) Workshop, June 2014 - August 2014
  • Investigated the use of multinomial subspace model iVectors, a dimension reduction technique, on phonotactic language recognition evaluation.

UIUC Statistical Speech Technology Group

Undergraduate Research Assistant, June 2012 - May 2014
  • Completed senior thesis on KL-Divergence Guided Two-Beam Viterbi Algorithm on Factorial Hidden Markov Models (HMMs).
  • Proposed and developed a variation of the Viterbi beam algorithm for more efficient decoding of HMMs.
  • Built a system in MATLAB for recognition of a pair of isolated digits on one audio channel.
  • Ran simulations, processed speech data in Matlab and Python.

Johns Hopkins Univ., Human Language Technology Center of Excellence

Research Intern, Summer Camp for Applied Language Exploration (SCALE) Workshop, June 2013 - August 2013
  • Paralleled and converted Matlab code into Python to maximize performance and the usage of the grid.
  • Profiled and optimized code in both Matlab and Python for better performance.

Teaching Experience

University of Illinois at Urbana-Champaign

Teaching Assistant, Fall 2016
  • ECE544NA: Pattern Recognition

University of Illinois at Urbana-Champaign

Teaching Assistant, Fall 2014, Spring 2015, Fall 2015
  • ECE420: Embedded DSP Laboratory

Skills

  • MATLAB, Python, C, C++, Java

Awards and Honors

Relevant Coursework

Graduate

  • CS543: Computer Vision
  • CS446: Machine Learning
  • ECE598MR: Statistical Learning Theory
  • ECE563: Information Theory
  • ECE551: Digital Signal Processing, II
  • ECE544NA: Pattern Recognition
  • ECE534: Random Processes

Undergraduate

  • CS440: Artificial Intelligence
  • CS411: Database Systems
  • CS373: Introduction to Theory of Computation
  • CS225: Data Structures
  • ECE496/499: Senior Thesis
  • ECE445: Senior Design
  • ECE420: Embedded DSP Laboratory
  • ECE418: Introduction to Image & Video Processing
  • ECE417: Multimedia Signal Processing
  • ECE310: Digital Signal Processing
  • MATH415: Linear Algebra





Undergraduate Projects


LED Cube Project

An LED cube is inherently aesthetically pleasing and ours will be capable of displaying 3D animations and lighting patterns with much increased complexity compared to any 2D display of comparable resolution.


AxWall Project

AxWall is a database driven website, where people can upload and share wallpapers that they have made or found. We accept user submissions, and allow users to browse for wallpaper by several categories; screen resolution, color, category, etc.


Multimodal Person Identification Android App.

Android Application for person identification using face and speech recognition. The applicaiton was able to identify a person with accuracy ~95% out of a database of five people.


Image Segmentation and Categorization

A month long project with the task to of class based pixel-wise segmentation and categorization using the single-histogram class model, and random forests.