I'm a computer vision engineer currently developing perception algorithms for autonomous vehicles. I have a particular interest in image-based object detection, passive sensing, and the remarkable capacity and efficacy at which humans perceive the natural world.

I previously studied at the Queensland University of Technology in Australia, under the supervision of Dr Simon Lucey. My research focussed on unsupervised image alignment and object registration. I have worked on a number of open source projects, notably the OpenCV library.

I take an avid interest in programming languages and language design, and along with being fluent in C++ and Python, my style is heavily influenced by functional programming and Common Lisp. My mathematical background is in linear algebra and optimization theory.

PhD Thesis

H. Bristow, Registration and Representation in Computer Vision, 2015.


H. Bristow, J. Valmadre and S. Lucey, Dense Semantic Correspondence Where Every Pixel is a Classifier, ICCV 2015.
H. Bristow and S. Lucey, In Defense of Gradient-Based Alignment on Densely Sampled Sparse Features, Springer Book on Dense Correspondences in Computer Vision, 2014.
H. Bristow and S. Lucey, Optimization Methods for Convolutional Sparse Coding, arXiv 2014
H. Bristow and S. Lucey, Why do Linear SVMs Trained on HOG Features Perform so Well, arXiv 2014
H. Bristow, A. Eriksson and S. Lucey, Fast Convolutional Sparse Coding, CVPR 2013
H. Bristow and S. Lucey, V1-Inspired Features Induce a Weighted Margin in SVMs, ECCV 2012