DocumentCode :
716867
Title :
Learning the shape of image moments for optimal 3D structure estimation
Author :
Giordano, Paolo Robuffo ; Spica, Riccardo ; Chaumette, Francois
Author_Institution :
Irisa & Inria Rennes Bretagne Atlantique, Rennes, France
fYear :
2015
fDate :
26-30 May 2015
Firstpage :
5990
Lastpage :
5996
Abstract :
The selection of a suitable set of visual features for an optimal performance of closed-loop visual control or Structure from Motion (SfM) schemes is still an open problem in the visual servoing community. For instance, when considering integral region-based features such as image moments, only heuristic, partial, or local results are currently available for guiding the selection of an appropriate moment set. The goal of this paper is to propose a novel learning strategy able to automatically optimize online the shape of a given class of image moments as a function of the observed scene for improving the SfM performance in estimating the scene structure. As case study, the problem of recovering the (unknown) 3D parameters of a planar scene from measured moments and known camera motion is considered. The reported simulation results fully confirm the soundness of the approach and its superior performance over more consolidated solutions in increasing the information gain during the estimation task.
Keywords :
closed loop systems; visual servoing; SfM schemes; closed-loop visual control; estimation task; image moments; integral region-based features; online shape optimisation; optimal 3D structure estimation; shape learning; structure from motion schemes; visual features; visual servoing community; Cameras; Convergence; Estimation; Observability; Polynomials; Three-dimensional displays; Visualization;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Robotics and Automation (ICRA), 2015 IEEE International Conference on
Conference_Location :
Seattle, WA
Type :
conf
DOI :
10.1109/ICRA.2015.7140039
Filename :
7140039
Link To Document :
بازگشت