Three-dimensional motion computation and object segmentation in a long sequence of stereo frames
International Journal of Computer Vision
Three-dimensional computer vision: a geometric viewpoint
Three-dimensional computer vision: a geometric viewpoint
Performance of optical flow techniques
International Journal of Computer Vision
Smoothness in Layers: Motion segmentation using nonparametric mixture estimation.
CVPR '97 Proceedings of the 1997 Conference on Computer Vision and Pattern Recognition (CVPR '97)
Normalized Cuts and Image Segmentation
CVPR '97 Proceedings of the 1997 Conference on Computer Vision and Pattern Recognition (CVPR '97)
CVPR '96 Proceedings of the 1996 Conference on Computer Vision and Pattern Recognition (CVPR '96)
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
Incorporating non-motion cues into 3D motion segmentation
Computer Vision and Image Understanding
Extraction and temporal segmentation of multiple motion trajectories in human motion
Image and Vision Computing
Learning spatio-temporal dependency of local patches for complex motion segmentation
Computer Vision and Image Understanding
Incorporating non-motion cues into 3d motion segmentation
ECCV'06 Proceedings of the 9th European conference on Computer Vision - Volume Part III
Hi-index | 0.00 |
We propose an EM approach combined with the modified separation matrix scheme to perform 3D motion segmentation of the image sequence which contains multiple moving objects. We observe that, given the detected features and their 2D optical flow, in most cases the objects or their flow are separated very well from each other in space. The separation matrix method modified by using normalized cuts achieves expected grouping results for these cases. However, when the objects are overlapped spatially but undergoing independent motions, such as Ullman's co-axial transparent cylinder demonstration, there will be no proper affinity to perform segmentation by that way. Pure underlying 3D motion becomes the only cue to segment the scene. We exploit the EM algorithm to deal with such difficult cases. The scheme is tested on vast number of synthetic image sequences. Results with real image sequence are also given.