Xiaozhi Qu

Au MATIS depuis 2013

Téléphone actuel : 33 1 43 98 80 00 + 7148

Contact : xiaozhi.qu(at)ign.fr


Doctorant
Axe de recherche ATOLL



Sujet de recherche

Precise location is one of key issues for many georeferenced applications in urban area? such as autonomous navigation, urban environment digitizing. The basic method to localization is GPS combined with INS, which suffer multipath, mask phenomenon and drift. Moreover, it is too expensive to be widely used in daily application. Image based localization is extensively researched in computer vision, however it also has drift problem in large area.This subject aims at studying a solution for precise localization in urban area based on georeferenced landmark database. Bundle adjustment and uncertainty analysis theory will be taken into account in our subject to improve the precision of localization.

Keywords: Landmark, Bundle adjustment, Uncertainty




 



Formation

Thèse en cours en traitement du signal et des images depuis 2013 (Université Paris-Est, École Doctorale MSTIC):
Landmark based localization: Detection and update of landmarks with uncertainty analysis
Dirigée par Nicolas Paparoditis, encadrement : Bahman Soheilian et Bertrand Cannelle.

2010–2012 M.E : Photogrammetry and romote sensing. State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, China.

2006-2010 B.E : Photogrammetry and Remote Sensing. School of Remote Sensing and Information Engineering, Wuhan University, China.



Parcours professionnel

2012-2013 Engineer : Map Department, Tencent Technology Co., Ltd



Publications

Articles de conférences avec comité de lecture

B. Soheilian, X. Qu, M. Brédif. Landmark based localization : LBA refinement using MCMC-optimized projections of RJMCMC-extracted road marks. IEEE Intelligent Vehicles Symposium (IV2016), Gothenburg, Sweden, 19-22 June 2016.

X. Qu, B. Soheilian, E. Habets, N. Paparoditis. Evaluation of SIFT and SURF for vision based localization. International Archives of Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. 41 (B3), pp 685-692, ISPRS Congress, Prague, 2016.

B. Soheilian, X. Qu, M. Brédif. Localisationa base d’amers visuels: Cartographie et mise en correspondance de marquages au sol et intégration dans LBA. in Actes de la journée Transports Intelligents. Congrès francophone de Reconnaissance des Formes et Intelligence Artificielle (RFIA), Clermont Ferrand, France, June 2016.

X. Qu, B. Soheilian, N. Paparoditis. Vehicle localization using mono-camera and geo-referenced traffic signs. IEEE Intelligent Vehicles Symposium, IV2015, pp.605-610, Seoul, South Korea, 28 June - 1st July 2015.



Site internet de la recherche à l'IGN