Curriculum vitae - Yannick Verdie

Probability and Statistics. - Pattern recognition and Machine learning ... Image and geometry techniques: ... Thesis topic: « Urban scene modeling from airborne data ». Keywords: ... June 2008. Major: Master in Network & Telecommunication.
1MB taille 6 téléchargements 181 vues
30 AVE LOUIS VULLIEMIN, 1005 LAUSANNE Cell-Phone ( 4 1 ) 7 8 6 6 8 5 4 1 4 email Y A N N I C K . V E R D I E @ E P F L . C H WebSite H T T P : / / Y A N N I C K . V E R D I E . O N L I N E . F R FRENCH, 30 YEARS OLD (28/06/1984)

Yannick Verdie Engineer (PhD) in I mage Proc essing and Ge ometry Pro cessing My research interests are in image processing for computer vision and 3D reconstruction processes.


Image and geometry techniques:

- Probability and Statistics

- Local feature extraction and matching (SIFT, SURF, etc.)

- Pattern recognition and Machine learning

- Image-based tracking (Kalman, Particle filter, etc.)

- Image processing and Computer vision

- Object detection and recognition

- Computer graphics

- Face and Gesture recognition

- Proficiency in C/C++ programming

- 3D reconstruction

- Optimization methods in vision and learning

- Mesh simplification and labeling

- Multi-Core/GPU programming for vision

- 3D point-set processing

SKILLS Computer Science: C, C++, Python, ASM, UML and Java, Java for mobile ( J2ME on Nokia, Android SDK and NDK) CUDA, OpenCV, PCL, Boost, Matlab, CGAL, OpenGL, OpenAL, Qt, Knowledge and skills on web languages and SQL database OS: Linux (Debian-based), Windows, Mac OSX.

Languages: French (mother tongue), English (fluent – 10years+), Japanese (JLPT N5)

RECENTLY: Postdoc in Computer Science, EPFL, Switzerland

2014 – now

I am working on augmented reality projects for the CERN (inside the particle generator ATLAS).

EDUCATION PhD in Computer Science, INRIA, France

October 2013

Thesis topic: « Urban scene modeling from airborne data » Keywords: Image processing, Geometry processing, Urban scene reconstruction, Scene understanding, 3D modeling, Marked Point Processes, LiDAR data, Multi-View Stereo data, Energy minimization, Markov Random Field Advisors: J.Zerubia and F.Lafarge - PhD supported by CORDI grant, Jury : L. Van Goul (ETHZ), N. J. Mitra (UCL), F. Lafarge (INRIA), P. Alliez (INRIA), J. Zerubia (INRIA), M. Ortner (ASTRIUM), J-P. Pons (ACUTE3D) Visitor PhD student (5 months), National Institute of Informatics, Tokyo, Japan Topic: 3D data enhancement for point cloud registration. Advisor: A. Sugimoto - Collaboration supported by the NII grant.

MS in Computer Science (option Thesis), Virginia Tech, USA

May 2010

Thesis topic: « Surface Gesture & Object Tracking on Tabletop Devices » Major: signal processing

Engineer School (Master level), Telecom SudParis, France Major: Master in Network & Telecommunication.

Cumulative GPA 4.0 / 4.0

June 2008


2014 – now

European EDUSAFE project: Work in collaboration with CERN on augmented reality project for technical assistance inside the ATLAS detector.

European MAGELLAN project: Work with other technical partners on Augmented reality games from panorama.

Developer CGAL, Geometry Factory, France •

2012 – 2013

CGAL is a library in computational geometry available to users in industry and academia. More than 200 customers worldwide have already chosen to invest in this library.

Conception and implementation of a shape extraction package based on efficient RANSAC algorithm for extracting primitives from unorganized point clouds.

Developer OpenCV 2.1.1, Google, USA •

Summer 2010

OpenCV is an image processing library used in research (more than 500 algorithms, downloaded more than 2

millions of times, 40 thousand active people on the project). •

Conception of a new GUI compatible with Mac OS, Linux and Windows.

Implementation of new image processing functionalities (from Matlab)

Graduate Research Assistant, HCI Laboratory, Virginia Tech •

2008 – 2010

Tangram: Develop an interface to interact with virtual and real tangrams mostly using Image processing

algorithms. •

MirrorTrack (Thesis subject): Create and implement algorithms detecting and tracking hands in 3D with basic

cameras thanks to hand's reflection. •

DrummerGame: In charge of the signal processing part of a massive strategy video game: process the sound input

as a command to control the characters.


MeshMantics Version 1.0 (Febuary 2013): Software for 3D mesh labeling and semantics.

(Acquired by a French company ~15000€) •

Kazoe Version 1.0 (September 2012): Software to count objects from large pictures.

Midori Version 1.0 (September 2012): Software to detect Trees from large 3D point clouds.

Nagare Version 1.0 (September 2012): Software to extract lineic structures from larges pictures.

IMPORTANT SCIENTIFIC PUBLICATIONS AND PRESENTATIONS - Verdie Y, Lafarge F, Alliez P. - LOD Generation for Urban Scenes ACM Trans. on Graphics, (TOG 2015, will be presented at SIGGRAPH 2015) - Verdie Y, Moo Yi K, Fua P, Lepetit V. - TILDE: A Temporally Invariant Learned DEtector Proceedings of the Computer Vision and Pattern Recognition (CVPR 2015), Boston, USA - Verdie Y, Lafarge F. - Detecting parametric objects in large scenes by Monte Carlo sampling International Journal of Computer Vision (IJCV 2013) - Verdie Y, Lafarge F. - Efficient Monte Carlo sampler for detecting parametric objects in large scenes Proceedings of the European Conference on Computer Vision (ECCV 2012), Firenze, Italy - Verdie Y, Lafarge F, Zerubia J. Generating compact meshes under planar constraints: an automatic approach for modeling buildings from aerial LiDAR, (ICIP 2011), Brussels, Belgium - Verdie Y, Fang B, Quek F, MirrorTrack - Tracking with Reflection - Comparison with Top-Down Approach, (ICMI-MLMI 2009), Boston, USA

Referees: Professional: Dr. P. Alliez, Senior researcher and member of the CGAL Editorial Board at Geometry Factory, France [email protected], (33) 4 92 38 76 77 Vadim Pisarevsky, Research scientist and software Willow Garage, USA [email protected] Academic: Dr. F. Lafarge, Junior researcher and PhD advisor, member of the team TITANE, INRIA, France [email protected], (33) 4 92 38 77 31 Dr. A. Sugimoto, Senior researcher and head of the team Sugimoto lab, NII, Japan [email protected], (81) 3 (4212) 2566 Dr. J. Zerubia, Senior researcher and PhD advisor, head of the team AYIN, INRIA, France [email protected], (33) 4 92 38 76 43