A Flexible New Technique for Camera Calibration
IEEE Transactions on Pattern Analysis and Machine Intelligence
IEEE Transactions on Pattern Analysis and Machine Intelligence
Structure and Motion from Line Segments in Multiple Images
IEEE Transactions on Pattern Analysis and Machine Intelligence
Bundle Adjustment - A Modern Synthesis
ICCV '99 Proceedings of the International Workshop on Vision Algorithms: Theory and Practice
Scene Modelling, Recognition and Tracking with Invariant Image Features
ISMAR '04 Proceedings of the 3rd IEEE/ACM International Symposium on Mixed and Augmented Reality
3D Modeling of Outdoor Environments by Integrating Omnidirectional Range and Color Images
3DIM '05 Proceedings of the Fifth International Conference on 3-D Digital Imaging and Modeling
Line-Based Structure from Motion for Urban Environments
3DPVT '06 Proceedings of the Third International Symposium on 3D Data Processing, Visualization, and Transmission (3DPVT'06)
Construction of Feature Landmark Database Using Omnidirectional Videos and GPS Positions
3DIM '07 Proceedings of the Sixth International Conference on 3-D Digital Imaging and Modeling
Line reconstruction from many perspective images by factorization
CVPR'03 Proceedings of the 2003 IEEE computer society conference on Computer vision and pattern recognition
A vehicle-borne urban 3-D acquisition system using single-row laser range scanners
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
| Hi-index | 0.02 |
This paper describes a method for reconstructing 3D straight lines in the road environment from an image sequence captured by an in-vehicle camera. An actual road environment includes many straight lines such as edges of buildings, utility poles and road markings. These 3D lines can be used effectively for understanding the road scene; our proposed method, therefore, aims to reconstruct the 3D lines using an in-vehicle camera. The camera motion and the 3D line parameters are estimated simultaneously by minimizing the reprojection errors of corresponding edge segments in the image sequence. In the road environment, a forward-looking in-vehicle camera has difficultly in establishing a large parallax for accurate estimation of parameters. The accuracy in estimating parameters is improved by using constraints defined on the basis of a general knowledge of the structures. In experiments on an actual road environment, the camera motion and the 3D line parameters have been estimated from detected edge segments. The accuracy of the 3D lines was evaluated by comparing the estimated position and reference positions.