AntoineLetouzey

3D reconstruction of dislocations

During my time at EPFL, I worked at the interface between two laboratories ; CVLab and LSME. I was in charge of designing a software that would assist the physisists in studying nano-scalled 3D curvilinear structures from a set of 2D images acquired through tomography in transmission electron microscopy. This project involves both computer vision and software usability.

Want to know more?


This software is designed to be used by physisits with no background in computer science (and even less in computer vision). Its main features are :
  • Semi-automated 2D detection of dislocations
  • Auto-Calibration of the multiple input images
  • 3D reconstruction and visualization of the dislocations

The code is almost fully written in python with bits of C++/OpenGL. The GUI is based on the Qt framework. Please have a look at the project page on EPFL website for download and more informations.

Demonstration


Other materials


IMC Poster


Hide this panel.

Weta Digital

Well, I worked on secret stuff... Some of it made it to the movies.

Learning Shape Topology

The purpose of this work is to sequentially learn the correct topology of a scene. Based on incomplete observations, such as a temporal sequence of meshes reconstructed from visual hulls, we can progressively build a template mesh that will fit the entiere sequence. Our method rely on the assumption that object have a fixed topology and do not physicaly merge. Hence every new topological evidence observed is added to the template.

Want to know more?


General idea


idea

Pipeline of the progressive shape algorithm


pipeline

Selected results


results


Other materials


CVPR Paper CVPR Poster


Hide this panel.

Scene Flow

In this work we study how to incorporate, in an efficient way, various constraints when estimating dense motion information over 3D surfaces from temporal variations of the intensity function in several images. Our primary motivation is to provide robust motion cues that can be directly used by an application, e.g. interactive applications, or that can be fed into more advanced tasks such as surface tracking or segmentation. The approach is however not limited to a specific scenario and applies to any application that can benefit from low level motion information. We explore the combination of heterogeneous photometric and geometric cues in an unified framework. We fuse 2D and 3D, sparse and dense information to constrain the estimation of the instantaneous 3D scene flow on a surface. This project uses the output of any multi-camera system, such as Grimage, which gives for each time step a set of calibrated images. Our framework merges constraints coming from 2D normal flow, sparse 2D features matching in the images, sparse 3D features matching on the surfaces and a regularization term to provide an accurate and smooth motion field over the surface.

Want to know more?


Here are some videos showing results on real datasets




Pipeline of our method for depht maps


pipeline


Other materials


BMVC Poster BMVC Abstract VMV Slides


Hide this panel.

Master thesis - Plane detection in images

This is the work I did during my research internship during my second year of Master in Toulouse. The purpose was to use line correspondances in image pairs to detect simultaneously every plans in a scene. Fisrt I adapted an existing technic based on generalized principal component analysis developped by Yi Ma used to find the number and parameters of subspaces defined by data points in a space of higher dimension. The results were nice but very sensitiv to noise. That's why we decided to devellop a new approach, we used a technic that fits well in the context of urban scenes. Using an estimation of the projection of the border between every pair of plans we were able to segment all lines used as input and then compute each plan's set of parameters.

Teaching

Publications

2012

Article dans une revue

res_1.png
titre
Flot de scène
auteur
Antoine Letouzey, Benjamin Petit, Edmond Boyer
article
Traitement du Signal, Lavoisier, 2012, 29 (3-5), pp.255-281. 〈10.3166/ts.29.255-281〉
Accès au texte intégral et bibtex
https://hal.inria.fr/hal-00746460/file/TS.pdf BibTex

Communication dans un congrès

res1.png
titre
Progressive Shape Models
auteur
Antoine Letouzey, Edmond Boyer
article
CVPR - Computer Vision and Patern Recognition - 2012, Jun 2012, Providence, United States. IEEE, pp.190-197, 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 〈10.1109/CVPR.2012.6247675〉
Accès au texte intégral et bibtex
https://hal.inria.fr/hal-00677506/file/cvpr.pdf BibTex
improve.png
titre
Modèles progressifs de forme
auteur
Antoine Letouzey, Edmond Boyer
article
CORESA - COmpression et REprésentation des Signaux Audiovisuels - 2012, May 2012, Lille, France. 2012
Accès au texte intégral et bibtex
https://hal.inria.fr/hal-00683549/file/coresa_PSM.pdf BibTex
pipeline.jpg
titre
Flot de scène à partir d'images couleur et de cartes de profondeur
auteur
Antoine Letouzey, Benjamin Petit, Edmond Boyer
article
RFIA 2012 - Reconnaissance des Formes et Intelligence Artificielle, Jan 2012, Lyon, France. pp.978-2-9539515-2-3, 2012
Accès au texte intégral et bibtex
https://hal.archives-ouvertes.fr/hal-00656484/file/rfia2012_submission_4.pdf BibTex

Thèse

titre
Modélisation 4D à partir de plusieurs caméras
auteur
Antoine Letouzey
article
Autre [cs.OH]. Université de Grenoble, 2012. Français. 〈NNT : 2012GRENM054〉
Accès au texte intégral et bibtex
https://tel.archives-ouvertes.fr/tel-00776075/file/24493_LETOUZEY_2012_archivage.pdf BibTex

2011

Communication dans un congrès

FlashKickSceneFlowCst0.png
titre
Surface Flow from Visual Cues
auteur
Benjamin Petit, Antoine Letouzey, Edmond Boyer, Jean-Sébastien Franco
article
Peter Eisert and Joachim Hornegger and Konrad Polthier. VMV 2011 - Vision, Modeling and Visualization Workshop, Oct 2011, Berlin, Germany. Eurographics Association, pp.1-8, 2011, Proceedings of the Vision, Modeling, and Visualization Workshop 2011. 〈http://diglib.eg.org/EG/DL/PE/VMV/VMV11/PDF/001-008.pdf〉. 〈10.2312/PE/VMV/VMV11/001-008〉
Accès au texte intégral et bibtex
https://hal.inria.fr/inria-00631476/file/surface_flow_from_visual_cues.pdf BibTex
overview.jpg
titre
Scene Flow from Depth and Color Images
auteur
Antoine Letouzey, Benjamin Petit, Edmond Boyer
article
Jesse Hoey and Stephen McKenna and Emanuele Trucco. BMVC 2011 - British Machine Vision Conference, Aug 2011, Dundee, United Kingdom. BMVA Press, pp.46:1-11, 2011, Proceedings of the British Machine Vision Conference. 〈10.5244/C.25.46〉
Accès au texte intégral et bibtex
https://hal.inria.fr/inria-00616353/file/main.pdf BibTex
BenAngeTrajectories.png
titre
Flot de surface à partir d'indices visuels
auteur
Benjamin Petit, Antoine Letouzey, Edmond Boyer
article
ORASIS - Congrès des jeunes chercheurs en vision par ordinateur, Jun 2011, Praz-sur-Arly, France. 2011
Accès au texte intégral et bibtex
https://hal.inria.fr/inria-00595247/file/orasis_id_11.pdf BibTex

Rapport

TeaserFinal.png
titre
Surface Flow from Visual Cues
auteur
Benjamin Petit, Antoine Letouzey, Edmond Boyer
article
[Research Report] RR-7619, INRIA. 2011, pp.18
Accès au texte intégral et bibtex
https://hal.inria.fr/inria-00593206/file/RR-7619.pdf BibTex