|
3D Immersive Visual Maps
In this work we have produced immersive 3D maps for real-time localisation and autonomous navigation In particular we have developped a method and apparatus for building dense visual maps of large scale environments for real-time localisation and autonomous navigation. We propose a spherical ego-centric representation of the environment which is able to reproduce photo-realistic omnidirectional views of captured environments. This representation is composed of a graph of locally accurate augmented spherical panoramas that allows to generate varying viewpoints through novel view synthesis. The spheres are related by a graph of 6 d.o.f. poses which are estimated through multi-view spherical registration. It is shown that this representation can be used to accurately localise a vehicle navigating within the spherical graph, using only a monocular camera for accurate localisation. To perform this task, an efficient direct image registration technique is employed. This approach directly exploits the advantages of the spherical representation by minimising a photometric error between a current image and a reference sphere. Autonomous navigation results are shown in challenging urban environments, containing pedestrians and other vehicles.
Example of an augmented spherical panorama obtained using a multi-baseline spherical camera:
CLICK ON THE IMAGE TO SEE THE VIDEO
The related publication can be found here:
Meilland, M., Rives, P. & Comport, A. I (2012). Dense RGB-D mapping for real-time localisation and navigation.. In IV12 Workshop on Navigation Positiong and Mapping. Alcalá de Henares, Spain.. [More] [Bibtex] [RIS] [MODS] |
|
Last Updated on Tuesday, 10 July 2012 17:34 |
|
|
We have recently been working on modelling illumination variation for direct dense localisation and mapping. Some of the associated publications with the following videos are:
[1] Meilland, M., Comport, A. I. & Rives, P (2011). Real-time Direct Model-based Tracking under Large Lighting Variations. In British Machine Vision Conference. University of Dundee, Scotland. [More] [Bibtex] [RIS] [MODS]
[2] Goncalves, T. & Comport, A. I (2011). Real-time Direct Tracking of Color Images in the Presence of Illumination Variation,. In IEEE International Conference on Robotics and Automation, ICRA'11. Shanghai, China. [More] [Bibtex] [RIS] [MODS]
- Link Video 1- Hybrid Illumination Tracking
- Link Video 2 - Local and Global illumination Model
| | |
