Harold Christopher Burger, PhD

A picture of myself.

I was formerly a PhD student at Max Planck Institute for Intelligent Systems, Tübingen, Germany, under the supervision of Bernhard Schölkopf and Stefan Harmeling, where my research focus was image denoising. I have achieved state-of-the-art results using neural networks.

I am currently co-founder at TasteHit, a web analytics and personalization startup. We aim to create a unique browsing experience tailored to a visitor's taste, using machine learning.

My CV

My email address.

Research

My primary research focus has been image denoising. I am particularly proud to have achieved state-of-the-art results. This was made possible through machine learning, and neural networks in particular. See the project page. You can also check out my page at the Max Planck Institute or my Google scholar page.

Software

This Matlab toolbox applies neural networks patch-wise for image denoising.
This method achieves the best denoising results reported in literature by a stand-alone method.
toolbox
Neural networks for various noise levels:
sigma=10 sigma=25 sigma=35 sigma=50 sigma=65 sigma=75 sigma=170 Warning: Large files!
NEW! (March 29, 2016): A python toolbox, kindly provided by Yuxiang Li.

This Matlab toolbox combines the results of the toolbox above and of BM3D, using neural networks.
For more details, see the paper "Learning how to combine internal and external denoising methods" and my PhD thesis.
This method achieves the best denoising results reported in literature by any method.
toolbox
Neural networks for various noise levels:
sigma=10 sigma=25 sigma=35 sigma=50 sigma=75 sigma=170 Warning: Large files!

Publications

Harold Christopher Burger, Christian J. Schuler, Stefan Harmeling
Learning how to combine internal and external denoising methods
pdf supplement web
@incproceedings{burgergcpr2013,
title={Learning how to combine internal and external denoising methods},
author={Burger, H.C. and Schuler, C.J. and Harmeling, S.},
booktitle={Proceedings of the 35th German Conference on Pattern Recognition (GCPR 2013)},
year={2013},
}

Harold Christopher Burger
Modelling and Learning Approaches to Image Denoising.
PhD thesis, University of Tübingen, 2013.
pdf slides
@phdthesis{burgerthesis2013,
title={Modelling and Learning Approaches to Image Denoising},
author={Burger, H.C.},
year={2013},
school={Eberhard Karls Universit\"{a}t T\"{u}bingen},
address = {Wilhelmstr. 32, 72074 T\"{u}bingen},
URL = {http://tobias-lib.uni-tuebingen.de/volltexte/2013/6821}
}

Christian J. Schuler, Harold Christopher Burger, Stefan Harmeling, Bernhard Schölkopf
A machine learning approach for non-blind image deconvolution. CVPR, 2013.
pdf supplement web
@inproceedings{schuler2013,
title={A machine learning approach for non-blind image deconvolution},
author={Schuler, C.J, and Burger, H.C. and Harmeling, S. and Sch\"{o}lkopf, B.},
booktitle={International Conference on Computer Vision and Pattern Recognition (CVPR)},
year={2013}
}

Mustafa Çavuşoğlu, Rolf Pohmann, Harold Christopher Burger, Kâmil Uludağ
Regional effects of magnetization dispersion on quantitative perfusion imaging for pulsed and continuous arterial spin labeling. Magnetic Resonance in Medicine, 2013.
pdf
@article{cavusoglu2012,
title={Regional effects of magnetization dispersion on quantitative perfusion imaging for pulsed and continuous arterial spin labeling},
author={Cavu{\c{s}}o{\u{g}}lu, M. and Pohmann, R. and Burger, H.C. and Uluda{\u{g}}, K.},
journal={Magnetic Resonance in Medicine},
volume={69},
issue={2},
pages={524--530},
year={2013}
}

Harold Christopher Burger, Christian J. Schuler, Stefan Harmeling.
Image denoising with multi-layer perceptrons, part 1: comparison with existing algorithms and with bounds. arXiv, 2012
pdf arXiv link
@article{burgerarxiv1,
author = {Burger, H. C. and Schuler, J. C. and Harmeling, S.},
title = {Image denoising with multi-layer perceptrons, part 1: comparison with existing algorithms and with bounds},
journal = {arXiv:1211.1544},
year = {2012}
}

Harold Christopher Burger, Christian J. Schuler, Stefan Harmeling
Image denoising with multi-layer perceptrons, part 2: training trade-offs and analysis of their mechanisms. arXiv, 2012.
pdf arXiv link
@article{burgerarxiv2,
author = {Burger, H. C. and Schuler, J. C. and Harmeling, S.},
title = {Image denoising with multi-layer perceptrons, part 2: training trade-offs and analysis of their mechanisms},
journal = {arXiv:1211.1552},
year = {2012}
}

Harold Christopher Burger, Christian J. Schuler, Stefan Harmeling
Image denoising: Can plain neural networks compete with BM3D? CVPR, 2012.
pdf supplement web poster
@inproceedings{burger2012,
title={Image denoising: Can plain neural networks compete with BM3D?},
author={Burger, H.C. and Schuler, C.J. and Harmeling, S.},
booktitle={International Conference on Computer Vision and Pattern Recognition (CVPR)},
year={2012}
}

Harold Christopher Burger, Stefan Harmeling
Improving denoising algorithms via a multi-scale meta-procedure. DAGM, 2011.
This paper was awarded with a prize at DAGM 2011.
pdf supplement poster
@inproceedings{burger2011a,
author = {Burger, H. C. and Harmeling, S.},
title = {Improving denoising algorithms via a multi-scale meta-procedure},
booktitle = {Proceedings of the 33rd international conference on Pattern recognition (DAGM)},
pages = {206--215},
year = {2011}
}

Harold Christopher Burger, Bernhard Schölkopf, Stefan Harmeling
Removing noise from astronomical images using a pixel-specific noise model. ICCP, 2011.
pdf supplement
@inproceedings{burger2011b,
title={Removing noise from astronomical images using a pixel-specific noise model},
author={Burger, H.C. and Sch\"{o}lkopf, B. and Harmeling, S.},
booktitle={International Conference on Computational Photography (ICCP)},
year={2011}
}

Christopher Malon, Matthew Miller, Harold Christopher Burger, Eric Cosatto, Hans Peter Graf
Identifying histological elements with convolutional neural networks. Proceedings of the 5th international conference on soft computing as transdisciplinary science and technology, 2008.
pdf
@inproceedings{malon2008,
title={Identifying histological elements with convolutional neural networks},
author={Malon, C. and Miller, M.L. and Burger, H.C. and Cosatto, E. and Graf, H.-P.},
booktitle={Proceedings of the 5th international conference on Soft computing as transdisciplinary science and technology},
year={2008}
}

Patents

Eric Cosatto, Harold Christopher Burger, Matthew L. Miller
Mitotic Figure Detector and Counter System and Method for Detecting and Counting Mitotic Figures.
US Patent App. 12/496,785, 2009.
link
@misc{cosatto2009, title={Mitotic Figure Detector and Counter System and Method for Detecting and Counting Mitotic Figures},
author={Cosatto, E. and Burger, H.C. and Miller, M.L.},
year={2009},
note={US Patent App. 12/496,785}
}