Development Engineer

HEATHER DUNLOP *******@*******.***

http://www.dunlop1.net

CAREER OBJECTIVE

A research and development position applying my expertise in computer vision.

ACADEMICS

Master of Science in Robotics, 2005 2006

Carnegie Mellon University, Pittsburgh, PA

GPA: 3.95/4.0

Advisor: David Wettergreen

Thesis: Automatic Rock Detection and Classification in Natural Scenes

Relevant Coursework: Projects:

Mathematical Foundations for Robotics Computing the Sphericity and Roundness of Rocks

Computer Vision A Statistical Approach to Texture Classification

Advanced Perception Detection & Segmentation of Rocks in MER Images

Computational Perception & Scene Analysis Rock Shape Characterization

Machine Learning Feature Selection for fMRI Classification

Bachelor of Mathematics, 2000 2005

University of Waterloo, Waterloo, ON

Honours Computer Science, Co-operative Program, Physics Minor

Overall Average: 82.9%, Computer Science Average: 86.3%

Nortel Networks Undergraduate Scholarship, 2000 2005

Relevant Coursework: Projects:

Computational Vision PCA for Template Searching and Matching

Computer Graphics Development of an extended ray tracer

Pattern Recognition

WORK EXPERIENCE

Digitalsmiths, Durham, NC

Image Computational Engineer, May 2007 present

R&D Team Lead, July 2009 present

Research, design, and implementation of image/video indexing and retrieval algorithms.

Projects include:

Scene classification of images and video (e.g., beach, forest, urban).

Face detection, tracking, and grouping.

Shot and scene boundary detection.

Clip retrieval using video signature.

Comparison of two versions of the same movie, e.g., rated vs. unrated (differential analysis).

Mars Space Flight Facility, Arizona State University, Tempe, AZ

Research Technician, Oct. 2006 Feb. 2007

Research and development of methods for the detection and size estimation of craters in satellite

images of Mars.

Development of a system to detect rocks in high-resolution Mars satellite imagery; formation of rock

distribution maps for selecting a landing site for the Phoenix Lander.

Robotics Institute, Carnegie Mellon University, Pittsburgh, PA

Research Assistant, Jan. Aug. 2006

Research, analysis, and development of methods for automatic detection and classification of rocks,

including: detection of rocks with an accurately delineated boundary, characterization of rock shape,

geologic classification with color, texture, and shape features.

Xiris Automation Inc., Burlington, ON

Software Developer/Tester, May Aug. 2004

Development in a team on applications for Ident code verification and printed graphics inspection

of optical discs.

Application testing, debugging, documentation.

Discreet Logic, Montreal, QC

Software Developer, Sept. Dec. 2003

C++ development in a Unix environment on a high performance visual effects package.

Integrated 3-D lookup table functionality into Flame.

Agris-Schoen Vision Systems, Inc., Alexandria, VA

Machine Vision Programmer, Sept. 2002 Apr. 2003

Assisted in software design process and wrote functional specifications.

Developed object-oriented software modules from design/functional specifications.

Evaluated signature matching algorithms.

Implemented machine vision feasibility studies on client samples sent for evaluation.

TECHNICAL SKILLS

Languages C/C++, Python, Matlab

Libraries OpenCV, FFmpeg, CUDA

Operating Systems Linux, Unix, Windows

Development Environments command-line development with GNU and Linux tools, version control

Development Skills object-oriented design, concurrency

PUBLICATIONS

H. Dunlop, Scene classification of images and video via semantic segmentation, in CVPR Workshop on

Perceptual Organization in Computer Vision, 2010.

H. Dunlop, D. Thompson, and D. Wettergreen, Multi-scale features for detection and segmentation of

rocks in mars images, in Proc. CVPR, 2007.

H. Dunlop, Automatic rock detection and classification in natural scenes, Master s thesis, Robotics Insti-

tute, Carnegie Mellon University, Pittsburgh, PA, August 2006.

J. Percy, H. Dunlop, L. Kassim, and R. Thompson, Periods of 25 pulsating red giants, International Astro-

nomical Union - Information Bulletin on Variable Stars, vol. 5041, 2001.

PATENTS

U.S. Patent Pending: Systems and Methods for Semantically Classifying Shots in Video, filed Feb. 2009.

Contact this candidate