Structure and Motion Estimation from Dynamic Silhouettes under Perspective Projection

Authors:
Tanuja Joshi;Narendra Ahuja;Jean Ponce
Affiliations:
Microsoft Corporation, One Microsoft Way, Redmond, WA 98052, USA;Beckman Institute, University of Illinois, Urbana, IL 61801, USA;Beckman Institute, University of Illinois, Urbana, IL 61801, USA
Venue:
International Journal of Computer Vision
Year:
1999

Citing 10
Cited 1

Solid shape

Solid shape
Using Extremal Boundaries for 3-D Object Modeling

IEEE Transactions on Pattern Analysis and Machine Intelligence - Special issue on interpretation of 3-D scenes—part II
Surface shape from the deformation of apparent contours

International Journal of Computer Vision
Disambiguating stereo matches with spatio-temporal surfaces

Geometric invariance in computer vision
3D Surface Reconstruction Using Occluding Contours

International Journal of Computer Vision
Acquiring 3-D Models from Sequences of Contours

IEEE Transactions on Pattern Analysis and Machine Intelligence
An Egomotion Algorithm Based on the Tracking of Arbitrary Curves

ECCV '92 Proceedings of the Second European Conference on Computer Vision
Robust Estimation of Surface Curvature from Deformation of Apparent Contours

ECCV '90 Proceedings of the First European Conference on Computer Vision
Relative 3D Regularized B-spline Surface Reconstruction Through Image Sequences

ECCV '94 Proceedings of the Third European Conference-Volume II on Computer Vision - Volume II
Motion from the frontier of curved surfaces

ICCV '95 Proceedings of the Fifth International Conference on Computer Vision

Epipolar Geometry from Profiles under Circular Motion

IEEE Transactions on Pattern Analysis and Machine Intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

We address the problem of estimating the structure and motion of asmooth curved object from its silhouettes observed over time by atrinocular stereo rig under perspective projection. We first constructa model for the local structure along the silhouette for each frame inthe temporal sequence. The local models are then integrated into aglobal surface description by estimating the motion between successivetime instants. The algorithm tracks certain surface features(parabolic points) and image features (silhouette inflections andfrontier points) which are used to bootstrap the motion estimationprocess. The entire silhouettes along with the reconstructed localstructure are then used to refine the initial motion estimate. Wehave implemented the proposed approach and report results on real images.