Accurate 3D car pose estimation

F. Chabot; M. Chaouch; J. Rabarisoa; C. Teuliere; T. Chateau

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

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

Document type :

Conference papers

Domain :

Engineering Sciences [physics] / Signal and Image processing

Complete list of metadatas

https://hal-cea.archives-ouvertes.fr/cea-01841166
Contributor : Léna Le Roy <>
Submitted on : Tuesday, July 17, 2018 - 10:15:47 AM
Last modification on : Wednesday, January 23, 2019 - 2:39:26 PM

Accurate 3D car pose estimation

Identifiers

Collections

Citation

Export

Metrics

234

Contact

Accurate 3D car pose estimation

Identifiers

Collections

Citation

Export

Share

Metrics

234

Contact