Region and Graph-Based Motion Segmentation

Authors:
Fernando C. Monteiro;Aurélio Campilho
Affiliations:
INEB - Instituto de Engenharia Biomédica, , and Instituto Politécnico de Bragança, , Portugal;INEB - Instituto de Engenharia Biomédica, , and FEUP - Faculdade de Engenharia, Universidade do Porto, Portugal
Venue:
ICIAR '08 Proceedings of the 5th international conference on Image Analysis and Recognition
Year:
2008

Citing 8
Cited 0

Empirical Bayesian Motion Segmentation

IEEE Transactions on Pattern Analysis and Machine Intelligence
A Unified Mixture Framework for Motion Segmentation: Incorporating Spatial Coherence and Estimating the Number of Models

CVPR '96 Proceedings of the 1996 Conference on Computer Vision and Pattern Recognition (CVPR '96)
Layered representation of motion video using robust maximum-likelihood estimation of mixture models and MDL encoding

ICCV '95 Proceedings of the Fifth International Conference on Computer Vision
Motion Segmentation and Tracking Using Normalized Cuts

ICCV '98 Proceedings of the Sixth International Conference on Computer Vision
Lucas/Kanade meets Horn/Schunck: combining local and global optic flow methods

International Journal of Computer Vision
Highly Accurate Optic Flow Computation with Theoretically Justified Warping

International Journal of Computer Vision
Spectral methods in image segmentation: a combined approach

IbPRIA'05 Proceedings of the Second Iberian conference on Pattern Recognition and Image Analysis - Volume Part II
Overview of the MPEG-7 standard

IEEE Transactions on Circuits and Systems for Video Technology

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper describes an approach for integrating motion estimation and region clustering techniques with the purpose of obtaining precise multiple motion segmentation. Motivated by the good results obtained in static segmentation we propose a hybrid approach where motion segmentation is achieved within a region-based clustering approach taken the initial result of a spatial pre-segmentation and extended to include motion information. Motion vectors are first estimated with a multiscale variational method applied directly over the input images and then refined by incorporating segmentation results into a region-based warping scheme. The complete algorithm facilitates obtaining spatially continuous segmentation maps which are closely related to actual object boundaries. A comparative study is made with some of the best known motion segmentation algorithms.