Email this article DETECTION OF MOVING VIDEO OBJECTS BASED ON LOCAL 3D-STRUCTURE TENSORS
- Authors: Favorskaya MN1
-
Affiliations:
- Issue: Vol 10, No 2 (2009)
- Pages: 141-146
- Section: Articles
- URL: https://journals.eco-vector.com/2712-8970/article/view/505020
- ID: 505020
Cite item
Full Text
Abstract
The analysis of video objects detection based on spatio-temporal approach is given. It covers the term of 3D-structure tensor for effective definition of spatio-temporal motion local orientation. The detailed schemes of pre-segmentation, spatio-temporal segmentation, and post-segmentation with further video objects recognition using local 3D-structure tensors are built.
References
- Bresson, X. A Variational Model for Object Segmentation Using Boundary Information and Shape Prior Driven by the Mumford-Shah Functional / Х. Bresson, P . V anderghey nst, J.-P . Thiran // Intern. J. of Computer V ision. 2006. V ol. 68, № 2. Р. 145-162.
- Cavallaro, A. Shadow-aware object-based video processing / А. Cavallaro, Е . Salvador, T. Ebrahimi // IEEE V ision, Image and Signal Proc. 2005. V ol. 152, Iss. 4. Р. 14-22.
- Thirde, D. Spatio-Temporal Semantic Object Segmentation using Probabilistic Sub-Object Regions / D. Thirde, G. Jones, J. Flack // British Machine V ision Conf. Norw ich, UK, 2003. P . 163-172.
- Wang, H.-Y. Spatio-Temporal Video Object Segmentation via Scale-Adaptive 3D Structure Tensor / H.-Y. Wang, K.-K. Ma // EURASIP : J. on Applied Signal Proc. 2004. V ol. 6. P . 798-813.
Supplementary files
