Accurate 3D car pose estimation

F. Chabot 1 M. Chaouch 1 J. Rabarisoa 1 C. Teuliere 2 T. Chateau 2
1 DIASI - Département Intelligence Ambiante et Systèmes Interactifs
2 IP - Institut Pascal
Abstract : We propose a new approach for accurate car pose estimation in images using only a dataset of 3D untextured models. Our algorithm detects both a car and its 3D pose. It is based on the matching of 3D models with the car in the image. With a part detector based on Convolutional Neural Networks, interest points corresponding to predefined 3D parts are extracted from the image. Then, we use the car geometry to find which parts are relevant across viewpoints. Finally, a 2D/3D pose estimator is used to recover the 3D pose of the car. The main contribution is to learn appearance and geometry models from 3D models dataset only. Experiments show that the method is very competitive for car detection and coarse viewpoint classification and improves the 3D pose estimation over the state-of-the-art methods.
Keywords : 3D pose estimation Car detection Convolutional neural network Geometry model New approaches Pose estimation State-of-the-art methods Image processing 3D models Object detection Neural networks Image matching Convolution
Type de document :
Communication dans un congrès
2016 IEEE International Conference on Image Processing (ICIP), Sep 2016, Phoenix, AZ, United States. IEEE Computer Society, 2016-August, pp.3807-3811, 2016, 〈10.1109/ICIP.2016.7533072〉
Liste complète des métadonnées
https://hal-cea.archives-ouvertes.fr/cea-01841166
Contributeur : Léna Le Roy <>
Soumis le : mardi 17 juillet 2018 - 10:15:47
Dernière modification le : mercredi 23 janvier 2019 - 14:39:26

Identifiants

Collections

CEA | INSTITUT_PASCAL | CEA-UPSAY | DRT | SIGMA-CLERMONT | PRES_CLERMONT | LIST | UNIV-PARIS-SACLAY

Citation

F. Chabot, M. Chaouch, J. Rabarisoa, C. Teuliere, T. Chateau. Accurate 3D car pose estimation. 2016 IEEE International Conference on Image Processing (ICIP), Sep 2016, Phoenix, AZ, United States. IEEE Computer Society, 2016-August, pp.3807-3811, 2016, 〈10.1109/ICIP.2016.7533072〉. 〈cea-01841166〉

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de la notice

171

Contact

support.ccsd.cnrs.fr
hal.support@ccsd.cnrs.fr
Logo CCSD