Many techniques have been proposed for segmenting feature point trajectories tracked through a video sequence into independent motions, but objects in the scene are usually assumed to undergo general 3-D motions. As a result, the segmentation accuracy considerably deteriorates in realistic video sequences in which object motions are nearly degenerate. In this paper, we propose a multi-stage unsupervised learning scheme first assuming degenerate motions and then assuming general 3-D motions and show by simulated and real video experiments that the segmentation accuracy significantly improves without compromising the accuracy for general 3-D motions.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Yasuyuki SUGAYA, Kenichi KANATANI, "Multi-Stage Unsupervised Learning for Multi-Body Motion Segmentation" in IEICE TRANSACTIONS on Information,
vol. E87-D, no. 7, pp. 1935-1942, July 2004, doi: .
Abstract: Many techniques have been proposed for segmenting feature point trajectories tracked through a video sequence into independent motions, but objects in the scene are usually assumed to undergo general 3-D motions. As a result, the segmentation accuracy considerably deteriorates in realistic video sequences in which object motions are nearly degenerate. In this paper, we propose a multi-stage unsupervised learning scheme first assuming degenerate motions and then assuming general 3-D motions and show by simulated and real video experiments that the segmentation accuracy significantly improves without compromising the accuracy for general 3-D motions.
URL: https://global.ieice.org/en_transactions/information/10.1587/e87-d_7_1935/_p
Copy
@ARTICLE{e87-d_7_1935,
author={Yasuyuki SUGAYA, Kenichi KANATANI, },
journal={IEICE TRANSACTIONS on Information},
title={Multi-Stage Unsupervised Learning for Multi-Body Motion Segmentation},
year={2004},
volume={E87-D},
number={7},
pages={1935-1942},
abstract={Many techniques have been proposed for segmenting feature point trajectories tracked through a video sequence into independent motions, but objects in the scene are usually assumed to undergo general 3-D motions. As a result, the segmentation accuracy considerably deteriorates in realistic video sequences in which object motions are nearly degenerate. In this paper, we propose a multi-stage unsupervised learning scheme first assuming degenerate motions and then assuming general 3-D motions and show by simulated and real video experiments that the segmentation accuracy significantly improves without compromising the accuracy for general 3-D motions.},
keywords={},
doi={},
ISSN={},
month={July},}
Copy
TY - JOUR
TI - Multi-Stage Unsupervised Learning for Multi-Body Motion Segmentation
T2 - IEICE TRANSACTIONS on Information
SP - 1935
EP - 1942
AU - Yasuyuki SUGAYA
AU - Kenichi KANATANI
PY - 2004
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E87-D
IS - 7
JA - IEICE TRANSACTIONS on Information
Y1 - July 2004
AB - Many techniques have been proposed for segmenting feature point trajectories tracked through a video sequence into independent motions, but objects in the scene are usually assumed to undergo general 3-D motions. As a result, the segmentation accuracy considerably deteriorates in realistic video sequences in which object motions are nearly degenerate. In this paper, we propose a multi-stage unsupervised learning scheme first assuming degenerate motions and then assuming general 3-D motions and show by simulated and real video experiments that the segmentation accuracy significantly improves without compromising the accuracy for general 3-D motions.
ER -