Since 2011 Software engineer at Parrot, Paris
Working on a solution for automatic speech recognition (ASR) and text-to-speech (TTS) in embedded systems. Development of the legacy C library used in several OEM and aftermarket products. Conception and implementation of the new version of the library written in C++ as a service in a custom Android 4.2 environment.
Developed tools in Python to generate the necessary binary files for various products. Created several Web applications to manage benchmarks and native speaker evaluations.
2009-11 Research Engineer at Commissariat à l'Energie Atomique (CEA), Saclay, France
Developed a library aimed at real time computer vision on massively parallel multi-core processors in embedded systems and graphics hardware.
Programs were written using an experimental framework which included a language (based on C89) that implemented dataflow principles to describe an application as a finite state machine.
Applications included multi-view pose estimation of a 3D model and human posture tracking and recognition, and real-time dense reconstruction..
2005-08 Ph.D. Candidate at Télécom Paris & INRIA, Rennes, France
Performed research in robust video compression of multi-terminal sources.
Built four video codecs based on a temporal two- or three-band motion-compensated temporal filtering with different different time splitting patterns and protected by a Wyner-Ziv stream.
Designed and studied an embedded multiple description scalar quantization scheme with side information for Gaussian sources.
Developed a low-complexity robust wavelet-based video encoding scheme based on multiple description coding with side information.
All codecs were written in C++ using the STL.
Advisors : Béatrice Pesquet-Popescu (ENST) and Christine Guillemot (INRIA).
2005 Research Intern at Télécom Paris, France
Developed an interpolation and decimation method for the HD video format that takes into account discontinuities like edges in the images by means of a non-linear decomposition using adaptive wavelets constructed on a lifting scheme with perfect reconstruction.
2003-04 Engineer Intern at Siemens Corporate Research, Princeton, USA
Performed analysis and comparisons in the field of compression applied to medical images.
Incorporated a C++ implementation of JPEG2000 3D (based on Kakadu) into Syngo, a software platform for medical imaging.
Evaluated the compressibility of multi-slice CT datasets, its dependence on slice thickness and the use of 3D vs. 2D JPEG2000 compression methods.
Studied the correlation of radiologists' image quality perception with quantitative assessment parameters: JND (Just Noticeable Differences) vs. PSNR metrics.
Compared objective image quality of CR scans using 8-bit and 11-bit DICOM images.
2001-02 Technical Intern at Bimsolutions, Paris, France
Developed scalable multi-tier mobile internet applications (Web, PDAs, WAP) using the Enhydra Java/XML application server.
Implemented a business software to optimize team management and simplify operational management with collaborators.
Designed and implemented a business software for mobile employees to report on their efforts and access corporate resources when working remotely.
2005-08 École Nationale Supérieure des Télécommunications (ENST), France
Ph.D. in Information Technology, Signal Processing.
2004-05 École Normale Supérieure de Cachan (ENS Cachan), France
M.Sc. in « Mathematics, Vision and Learning » (MVA), with honors.
2000-04 Université de Technologie de Compiègne (UTC), France
B.Sc. in Computer Engineering. Exchange student at the National University of Singapore (NUS) for a full academic year.
Automatic speech recognition and text-to-speech technologies (Nuance Vocon Hybrid, SVOX Automotive TTS), signal processing, image & video coding standards: JPEG2000, MPEG-4, 3D computer graphics, embedded systems, GPGPU, massively parallel architectures.
vim, Visual Studio, Matlab, Eclipse, CVS, SVN, Mercurial, Git, gcc, make, CMake, LaTeX.
French (mother tongue), English (fluent, 1 year in the United States and 1 year in Singapore), Spanish (intermediate), Mandarin (good notions).
Olivier Crave, “Approches théoriques en codage vidéo robuste multi-terminal”, Ph.D. Thesis, École Nationale Supérieure des Télécommunications, December 2008. Download: http://pastel.archives-ouvertes.fr/pastel-00004774/en/
Olivier Crave, Béatrice Pesquet-Popescu, Christine Guillemot, “Robust video coding based on multiple description scalar quantization with side information”, IEEE Transactions on Circuits and Systems for Video Technology, June 2010.
Olivier Crave, Christine Guillemot, Béatrice Pesquet-Popescu, “Multiple description coding with side information: Practical scheme and iterative decoding”, EURASIP Journal on Advances in Signal Processing, May 2009.
Olivier Crave, Christophe Tillier, Christine Guillemot, Béatrice Pesquet-Popescu, “Distributed Temporal Multiple Description Coding for Robust Video Transmission”, EURASIP Journal on Wireless Communications and Networking, Special Issue on Multimedia over Wireless Networks, Jan. 2008.
Olivier Crave, Béatrice Pesquet-Popescu, Christine Guillemot, “Multiple description video coding and iterative decoding of LDPCA codes with side information”, IEEE ICASSP, 34th International Conference on Acoustics, Speech, and Signal Processing, Taiwan, Apr. 2009.
Ҫağatay Dikici, Thomas Maugey, Marie Andreé Agostini, Olivier Crave, “Efficient frame interpolation for Wyner-Ziv video coding”, IS&T/SPIE Electronic Imaging, San Jose, Jan. 2009.
Olivier Crave, Christine Guillemot, Béatrice Pesquet-Popescu, “Multiple description source coding with side information”, EUSIPCO 2008, 16th European Signal Processing Conference, Lausanne, Aug. 2008.
Olivier Crave, Christine Guillemot, Béatrice Pesquet-Popescu, Christophe Tillier, “Robust video transmission based on distributed multiple description coding”, EUSIPCO 2007, 15th European Signal Processing Conference, Poznań, Sep. 2007.
Christophe Tillier, Olivier Crave, Béatrice Pesquet-Popescu, Christine Guillemot, “A comparison of four video multiple description coding schemes”, EUSIPCO 2007, 15th European Signal Processing Conference, Poznań, Sep. 2007.
Olivier Crave, Gemma Piella, Béatrice Pesquet-Popescu, “Image interpolation using an adaptive invertible approach”, EUSIPCO 2006, 14th European Signal Processing Conference, Florence, Sep. 2006.
IEEE Signal Processing Letters, IEEE Transactions on Circuits and Systems for Video Technology, EURASIP Journal on Wireless Communications and Networking, ICME, EUSIPCO, Bentham Recent Patents on Computer Science, Elsevier Signal Processing.
A tool for viewing stereolithography files: http://sourceforge.net/projects/stlviewer/
A rapid game prototyping framework: http://oboy.googlecode.com/
Cultural tourism, video games (www.cravesoft.com), reading, web edition, running.