Details
| Original language | English |
|---|---|
| Pages (from-to) | 73-80 |
| Number of pages | 8 |
| Journal | ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| Volume | 4 |
| Issue number | 2 |
| Publication status | Published - 28 May 2018 |
| Event | 2018 ISPRS TC II Mid-term Symposium "Towards Photogrammetry 2020" - Riva del Garda, Italy Duration: 4 Jun 2018 → 7 Jun 2018 |
Abstract
The precise reconstruction and pose estimation of vehicles plays an important role, e.g. for autonomous driving. We tackle this problem on the basis of street level stereo images obtained from a moving vehicle. Starting from initial vehicle detections, we use a deformable vehicle shape prior learned from CAD vehicle data to fully reconstruct the vehicles in 3D and to recover their 3D pose and shape. To fit a deformable vehicle model to each detection by inferring the optimal parameters for pose and shape, we define an energy function leveraging reconstructed 3D data, image information, the vehicle model and derived scene knowledge. To minimise the energy function, we apply a robust model fitting procedure based on iterative Monte Carlo model particle sampling. We evaluate our approach using the object detection and orientation estimation benchmark of the KITTI dataset (Geiger et al., 2012). Our approach can deal with very coarse pose initialisations and we achieve encouraging results with up to 82 % correct pose estimations. Moreover, we are able to deliver very precise orientation estimation results with an average absolute error smaller than 4°.
Keywords
- 3D Modelling, 3D Reconstruction, Active Shape Model, Object Detection, Pose Estimation, Stereo Images
ASJC Scopus subject areas
- Earth and Planetary Sciences(all)
- Earth and Planetary Sciences (miscellaneous)
- Environmental Science(all)
- Environmental Science (miscellaneous)
- Physics and Astronomy(all)
- Instrumentation
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol. 4, No. 2, 28.05.2018, p. 73-80.
Research output: Contribution to journal › Conference article › Research › peer review
}
TY - JOUR
T1 - Recovering the 3D pose and shape of vehicles from stereo images
AU - Coenen, M.
AU - Rottensteiner, F.
AU - Heipke, C.
N1 - Funding Information: This work was supported by the German Research Foundation (DFG) as a part of the Research Training Group i.c.sens [GRK2159].
PY - 2018/5/28
Y1 - 2018/5/28
N2 - The precise reconstruction and pose estimation of vehicles plays an important role, e.g. for autonomous driving. We tackle this problem on the basis of street level stereo images obtained from a moving vehicle. Starting from initial vehicle detections, we use a deformable vehicle shape prior learned from CAD vehicle data to fully reconstruct the vehicles in 3D and to recover their 3D pose and shape. To fit a deformable vehicle model to each detection by inferring the optimal parameters for pose and shape, we define an energy function leveraging reconstructed 3D data, image information, the vehicle model and derived scene knowledge. To minimise the energy function, we apply a robust model fitting procedure based on iterative Monte Carlo model particle sampling. We evaluate our approach using the object detection and orientation estimation benchmark of the KITTI dataset (Geiger et al., 2012). Our approach can deal with very coarse pose initialisations and we achieve encouraging results with up to 82 % correct pose estimations. Moreover, we are able to deliver very precise orientation estimation results with an average absolute error smaller than 4°.
AB - The precise reconstruction and pose estimation of vehicles plays an important role, e.g. for autonomous driving. We tackle this problem on the basis of street level stereo images obtained from a moving vehicle. Starting from initial vehicle detections, we use a deformable vehicle shape prior learned from CAD vehicle data to fully reconstruct the vehicles in 3D and to recover their 3D pose and shape. To fit a deformable vehicle model to each detection by inferring the optimal parameters for pose and shape, we define an energy function leveraging reconstructed 3D data, image information, the vehicle model and derived scene knowledge. To minimise the energy function, we apply a robust model fitting procedure based on iterative Monte Carlo model particle sampling. We evaluate our approach using the object detection and orientation estimation benchmark of the KITTI dataset (Geiger et al., 2012). Our approach can deal with very coarse pose initialisations and we achieve encouraging results with up to 82 % correct pose estimations. Moreover, we are able to deliver very precise orientation estimation results with an average absolute error smaller than 4°.
KW - 3D Modelling
KW - 3D Reconstruction
KW - Active Shape Model
KW - Object Detection
KW - Pose Estimation
KW - Stereo Images
UR - http://www.scopus.com/inward/record.url?scp=85048426161&partnerID=8YFLogxK
U2 - 10.5194/isprs-annals-IV-2-73-2018
DO - 10.5194/isprs-annals-IV-2-73-2018
M3 - Conference article
AN - SCOPUS:85048426161
VL - 4
SP - 73
EP - 80
JO - ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
JF - ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
SN - 2194-9042
IS - 2
T2 - 2018 ISPRS TC II Mid-term Symposium "Towards Photogrammetry 2020"
Y2 - 4 June 2018 through 7 June 2018
ER -