A Closed-Form Solution to Non-Rigid Shape and Motion Recovery
International Journal of Computer Vision
Two-View Multibody Structure from Motion
International Journal of Computer Vision
A Unified Algebraic Approach to 2-D and 3-D Motion Segmentation and Estimation
Journal of Mathematical Imaging and Vision
Motion segmentation using GPCA techniques and optical flow
EATIS '07 Proceedings of the 2007 Euro American conference on Telematics and information systems
Generalized principal component analysis (GPCA)
CVPR'03 Proceedings of the 2003 IEEE computer society conference on Computer vision and pattern recognition
Direct segmentation of multiple 2-D motion models of different types
WDV'05/WDV'06/ICCV'05/ECCV'06 Proceedings of the 2005/2006 international conference on Dynamical vision
A bottom up algebraic approach to motion segmentation
ACCV'06 Proceedings of the 7th Asian conference on Computer Vision - Volume Part I
| Hi-index | 0.00 |
We present an algebraic geometric approach for segmentingboth static and dynamic scenes from image intensities.We introduce the multibody affine constraint as ageometric relationship between the motion of multiple objectsand the image intensities generated by them. This constraintis satisfied by all the pixels, regardless of the body towhich they belong and regardless of depth discontinuities orperspective effects. We propose a polynomial factorizationtechnique that estimates the number of affine motion modelsas well as their motion parameters in polynomial time.The factorization technique is used to initialize a nonlinearalgorithm that minimizes the algebraic error defined by themultibody affine constraint. Our approach is based solelyon image intensities, hence it does not require feature trackingor correspondences. It is therefore a natural generalizationof the so-called direct methods in single-body structurefrom motion to multiple moving objects. We present simulationand experimental results that validate our approach.