Efficient City-Sized 3D Reconstruction from Ultra-High Resolution Aerial and Ground Video Imagery

This paper introduces an approach for geo-registered, dense 3D reconstruction of city-sized scenes using a combination of ultra-high resolution aerial and ground video imagery. While 3D reconstructions from ground imagery provide high-detail street-level views of a city, they do not completely cover the entire city scene and might have distortions due to GPS drift. Such a reconstruction can be complemented by aerial imagery to capture missing scene surfaces as well as improve geo-registration. We present a computationally efficient method for 3D reconstruction of city-sized scenes using both aerial and ground video imagery to obtain a more complete and self-consistent georegistered 3D city model. The reconstruction results of a 1×1km city area, covered with a 66 Mega-pixel airborne system along with a 60 Mega-pixel ground camera system, are presented and validated to geo-register to within 3m to prior airborne-collected LiDAR data.

[1]  Paul J. Besl,et al.  A Method for Registration of 3-D Shapes , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[2]  Marc Levoy,et al.  Efficient variants of the ICP algorithm , 2001, Proceedings Third International Conference on 3-D Digital Imaging and Modeling.

[3]  David Nistér,et al.  An efficient solution to the five-point relative pose problem , 2004, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4]  David G. Lowe,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004, International Journal of Computer Vision.

[5]  Christian Früh,et al.  An Automated Method for Large-Scale, Ground-Based City Model Acquisition , 2004, International Journal of Computer Vision.

[6]  Kurt Konolige,et al.  Real-time Localization in Outdoor Environments using Stereo Vision and Inexpensive GPS , 2006, 18th International Conference on Pattern Recognition (ICPR'06).

[7]  Naokazu Yokoya,et al.  Extrinsic Camera Parameter Estimation Based-on Feature Tracking and GPS Data , 2006, ACCV.

[8]  Marko Modsching,et al.  Field trial on GPS Accuracy in a medium size city: The influence of built- up 1 , 2006 .

[9]  Jan-Michael Frahm,et al.  Detailed Real-Time Urban 3D Reconstruction from Video , 2007, International Journal of Computer Vision.

[10]  Zuzana Kukelova,et al.  3D reconstruction from image collections with a single known focal length , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[11]  Taehee Lee,et al.  Robust 3D street-view reconstruction using sky motion estimation , 2009, 2009 IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops.

[12]  Andrew Zisserman,et al.  Multiple View Geometry , 1999 .

[13]  Richard Szeliski,et al.  Building Rome in a day , 2009, ICCV.

[14]  Jana Kosecka,et al.  Piecewise planar city 3D modeling from street view panoramic sequences , 2009, CVPR.

[15]  Jean Ponce,et al.  Accurate, Dense, and Robust Multiview Stereopsis , 2010, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[16]  Marc Pollefeys,et al.  Fast robust large-scale mapping from video and internet photo collections , 2010 .

[17]  Pascal Fua,et al.  Dynamic and scalable large scale image reconstruction , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[18]  Jan-Michael Frahm,et al.  Building Rome on a Cloudless Day , 2010, ECCV.

[19]  Thomas Deselaers,et al.  ClassCut for Unsupervised Class Segmentation , 2010, ECCV.

[20]  Noah Snavely,et al.  Scene Reconstruction and Visualization from Internet Photo Collections: A Survey , 2011, IPSJ Trans. Comput. Vis. Appl..