(original) (raw)

COMPETITIONS

Handwritten Chinese Characters (ICDAR 2011 & 2013)

Chinese characters

Traffic Sign Recognition (IJCNN 2012)

traffic signs

Neuronal Membrane Segmentation (ISBI 2012)

traffic signs

Mitosis Detection

(ICPR 2012 & MICCAI 2013)

mitosis detection

On February 1, 2017 I will quit my position at IDSIA to focus on my startup.

Dan Claudiu Cireșan

Senior Researcher - Dalle Molle Institute for Artificial Intelligence (IDSIA)

Interest

AI, Machine Learning, Pattern Recognition, Deep Learning with Neural Networks
Computer Vision
Document Analysis
Speech Recognition
Medical and Biological Image Analysis (joint work with Alessandro Giusti)
Automotive
GPGPU

News

April 5 2016: NVIDIA awarded IDSIA with a DGX-1 supercomputer. My contribution is research with GPUs for Medical Applications, Transfer Learning and Deep Learning.
Our research with autonomous quadcopter has been featured in NVIDIA's Introduction of Opening Keynote from GTC 2016.
I will present a tutorial at VISUM 2016. Join VISion Understanding and Machine intelligence Summer School. Apply before March 21st.
See how my Deep Neural Network drives a quadcopter along forest trails in our latest work: video, paper. More info on Alessandro's webpage.

Deep Neural Networks for Academic Projects

Do you have an idea which can benefit from my Deep Neural Networks? I am looking for research partners for projects (SNSF, CTI, EU, IARPA/DARPA, etc) related to the topics listed above.

Deep Neural Networks for Industry

Competitions (all methods use my Deep NN framework)

First place at Assessment of Mitosis Detection Algorithms, MICCAI 2013 Grand Challenge, Nagoya, Japan (with Alessandro Giusti).
Best score on test set from Chinese Handwriting Recognition Competition; task: offline characters, ICDAR 2013, Dallas, US - details in Multi-Column Deep Neural Networks for Offline Handwritten Chinese Character Classification - IDSIA Technical Report, August 2013.
First place at Mitosis Detection in Breast Cancer Histological Images, ICPR 2012, Tsukuba, Japan (with Alessandro Giusti).
First place at Segmentation of neuronal structures in EM stacks challenge - ISBI 2012, Barcelona, Spain (with Alessandro Giusti). We were the only team with better than human pixel level segmentation performance.
First place at Offline Chinese Character Recognition (task1: "Offline Chinese Character Recognition") at ICDAR 2011, Beijing, China (with Ueli Meier).
First place at The German Traffic Sign Recognition Benchmark (both phases) at IJCNN 2011, San Jose, US (with Ueli Meier and Jonathan Masci). We were the only team with better than human performance.

Talks

Link�ping University, October 2015
NVIDIA Round Table Meeting, Bonn, September 2015
Talk at TEDxBrussels featuring several of my results.
plenary talk at International Work Conference on Artificial Neural Networks (IWANN 2015), Palma de Mallorca, June 2015. Slides: plenary talk, transfer learning
ITN Marie Curie Joint Summer School CONTEST - PROTOTOUCH, May 2015.
GPU Technology Conference (GTC 2015), San Jose, March 2015. Video of my talk.
Redwood Center for Theoretical Neuroscience, Berkeley, November 2014
NVIDIA, SC 2014, November 17-20, New Orleans, Louisiana
ETH, Zurich, October 2014
NVIDIA Round Table Meeting, Bonn, September 2014
NVIDIA Webinar Deep Neural Networks for Visual Pattern Recognition, August 2014. Answers to some of the questions I got during the webinar.
Summer School on Deep Learning for Image Analysis, Langeland, Denmark, 2014
Institut f�r Neuro- und Bioinformatik, Lübeck, May 2014
CUDA Spotlight, April 2014
Deepmind, London, December 2012
Janelia Farm, HHMI, Ashburn, June 2012
Department of Brain and Cognitive Sciences, MIT, Boston, June 2012
Siemens Corporate Research, Princeton, August 2011

Publications Google Scholar citations dblp linkedin

Journal papers:

NEW! A. Giusti, J. Guzzi, D. Ciresan, F. Lin He, J. P. Rodriguez, F. Fontana, M. Faessler, C. Forster, J. Schmidhuber, G. A. Di Caro, D. Scaramuzza, L. Gambardella - A Machine Learning Approach to Visual Perception of Forest Trails for Mobile Robots (IEEE Robotics and Automation Letters, 2015, paper, bib)
I. Arganda-Carreras et al. - Crowdsourcing the creation of image segmentation algorithms for connectomics (Frontiers in Neuroanatomy, 2015)
M. Veta et al. - Assessment of algorithms for mitosis detection in breast cancer histopathology images (Medical Image Analysis, 2015, paper, bib)
D. Ciresan, U. Meier, J. Masci, J. Schmidhuber - Multi Column Deep Neural Network for Traffic Sign Classification (invited, Neural Networks 2012, bib, paper, http://dx.doi.org/10.1016/j.neunet.2012.02.023)
D. Ciresan, U. Meier, L. M. Gambardella, J. Schmidhuber - Deep, Big, Simple Neural Nets for Handwritten Digit Recognition (Neural Computation, December 2010, bib, paper) Book chapters:
D. C. Ciresan, U. Meier, L. M. Gambardella, J. Schmidhuber - Deep Big Multilayer Perceptrons For Digit Recognition (Neural Networks Tricks of the Trade, Springer, 2012, bib, preview) Conference papers:

2015

P. Toufiq, D. Ciresan, A. Giusti - Efficient Classifier Training to Reduce False Merges in Electron Microscopy Segmentation (ICCV 2015, paper, bib)
D. Ciresan, U. Meier - Multi-Column Deep Neural Networks for offline handwritten Chinese character classification (IJCNN 2015, bib) 2014
J. Funke, J. Martel, S. Gerhard, B. Andres, D. Ciresan, A. Giusti, L.M. Gambardella, J. Schmidhuber, H. Pfister, A. Cardona, M. Cook - Candidate Sampling for Neuron Reconstruction from Anisotropic Electron Microscopy Volumes (MICCAI 2014, bib, paper preview)
A. Giusti, C. Caccia, D. Ciresan, J. Schmidhuber, L.M. Gambardella - A Comparison of Algorithms and Humans for Mitosis Detection (ISBI 2014, bib, paper preview) 2013
D. Ciresan, A. Giusti, L.M. Gambardella, J. Schmidhuber - Mitosis Detection in Breast Cancer Histology Images using Deep Neural Networks (MICCAI 2013, bib, paper preview, poster)
A. Giusti, D. Ciresan, J. Masci, L.M. Gambardella, J. Schmidhuber - Fast Image Scanning with Deep Max-Pooling Convolutional Neural Networks (ICIP 2013, bib, paper preview)
J. Masci, A. Giusti, D. Ciresan, G. Fricout, J. Schmidhuber - A Fast Learning Algorithm for Image Segmentation with Max-Pooling Convolutional Networks (ICIP 2013, bib, paper preview) 2012
D. Ciresan, A. Giusti, L. M. Gambardella, J. Schmidhuber - Deep Neural Networks Segment Neuronal Membranes in Electron Microscopy Images (NIPS 2012, bib, paper, poster, results)
D. Ciresan, U. Meier, J. Schmidhuber - Multi-column Deep Neural Networks for Image Classification (CVPR 2012, bib, paper, supplementary material, poster, erratum)
D. Ciresan, U. Meier, J. Schmidhuber - Transfer Learning for Latin and Chinese Characters with Deep Neural Networks (IJCNN 2012, bib, paper)
J. Masci, U. Meier, D. Ciresan, J. Schmidhuber - Steel Defect Classification with Max-Pooling Convolutional Neural Networks (IJCNN 2012, bib, paper) 2011
D. Ciresan, U. Meier, L. M. Gambardella, J. Schmidhuber - Convolutional Neural Network Committees For Handwritten Character Classification (ICDAR 2011, bib, paper)
U. Meier, D. Ciresan, L. M. Gambardella, J. Schmidhuber - Better digit recognition with a committee of simple Neural Nets (ICDAR 2011, bib, paper)
D. Ciresan, U. Meier, J. Masci, L. M. Gambardella, J. Schmidhuber - Flexible, High Performance Convolutional Neural Networks for Image Classification (IJCAI 2011, bib, paper, slides, poster, video)
D. Ciresan, U. Meier, J. Masci, J. Schmidhuber - A Committee of Neural Networks for Traffic Sign Classification (IJCNN 2011, bib, paper, conference slides, workshop slides)
J. Masci, U. Meier, D. Ciresan, J. Schmidhuber - Stacked Convolutional Auto-Encoders for Hierarchical Feature Extraction (ICANN 2011, bib, paper, slides)
J. Schmidhuber, D. Ciresan, U. Meier, J. Masci, Alex Graves - On Fast Deep Nets for AGI Vision (AGI 2011, bib, paper, talk)
J. Nagi, F. Ducatelle, G. A. Di Caro, D. Ciresan, U. Meier, A. Giusti, F. Nagi, J. Schmidhuber, L. M. Gambardella - Max-Pooling Convolutional Neural Networks for Vision-based Hand Gesture Recognition (ICSIPA 2011, bib, paper) 2008
D. Ciresan - Avoiding Segmentation in Multi-Digit Numeral String Recognition by Combining Single and Two-Digit Classifiers Trained without Negative Examples (SYNASC 2008, bib, paper)
D. Ciresan, D. Pescaru - Off-line Recognition of Handwritten Numeral Strings Composed from Two-digits Partially Overlapped Using Convolutional Neural Networks (ICCP 2008)

Workshop papers:

D. Ciresan, A. Giusti, L. Gambardella, J. Schmidhuber - Neural Networks for Segmenting Neuronal Structures in EM Stacks (ISBI CH2 2012, bib, [paper](data/ISBI2012.pdf "A pixel classifier is driving our winning approach to neural structure segmentation in electron microscopy images. Without explicit feature computation, it predicts the label of each pixel (membrane or non-membrane) from raw pixel values in a square window centered on it. The classifier is a special type of feed-forward neural network trained by plain gradient descent. The input layer maps each window pixel to a neuron. It is followed by a succession of convolutional and max-pooling layers which preserve 2D information and extract features with increasing levels of abstraction. The output layer produces a calibrated probability for each class, and is subjected to very mild post-processing. The approach outperforms all competing entries in all considered metrics: values for \emph{rand error}, \emph{warping error} and \emph{pixel error} were 48⋅10−348\cdot 10^{-3}48⋅10−3, 434⋅10−6434\cdot 10^{-6}434⋅10−6 and 60⋅10−360\cdot 10^{-3}60⋅10−3, respectively."). Please cite the detailed NIPS 2012 paper)

Technical reports:

Multi-Column Deep Neural Networks for Offline Handwritten Chinese Character Classification - D. Ciresan, J. Schmidhuber, IDSIA Technical Report, August 2013
A Fast Learning Algorithm for Image Segmentation with Max-Pooling Convolutional Networks - J. Masci, A. Giusti, D. Ciresan, G. Fricout, J. Schmidhuber, IDSIA Technical Report, January 2013, accepted at ICIP 2013
Fast Image Scanning with Deep Max-Pooling Convolutional Neural Networks - A. Giusti, D. Ciresan, J. Masci, L.M. Gambardella, J. Schmidhuber, IDSIA Technical Report, January 2013, accepted at ICIP 2013
MCDNN for Image Classification (IDSIA Technical Report, February 2012, presented at CVPR 2012)
Handwritten Digit Recognition with a Committee of Deep Neural Nets on GPUs (IDSIA Technical Report, March 2011, published in Neural Networks Tricks of the Trade, Springer, 2012)
High-Performance Neural Networks for Visual Object Classification (IDSIA Technical Report, January 2011, presented at IJCAI 2011)

Projects

industry projects
Supervised Deep / Recurrent Nets, SNF grant 140399
Swiss CTI, Commission for Technology and Innovation, Project n. 9688.1 IFF: Intelligent Fill in Form (2009-2010)
Neural Dynamics, EU project

Current work

speech recognition
using NNs to solve computer vision tasks in robotics, medicine, general 3D scene understanding
connectomics
cancer detection
visual data mining
improving the GPU NN framework