Please use this identifier to cite or link to this item:
|Scopus||Web of Science®||Altmetric|
|Title:||Efficient search methods and deep belief networks with particle filtering for non-rigid tracking: application to lip tracking|
|Citation:||Proceedings of the 17th IEEE International Conference on Image Processing, held in Hong Kong, 26-29 September, 2010: pp.3817-3820|
|Publisher:||IEEE Computer Society|
|Series/Report no.:||IEEE International Conference on Image Processing ICIP|
|Conference Name:||IEEE International Conference on Image Processing (17th : 2012 : Hong Kong)|
|Jacinto C. Nascimento and Gustavo Carneiro|
|Abstract:||Pattern recognition methods have become a powerful tool for segmentation in the sense that they are capable of automatically building a segmentation model from training images. However, they present several difficulties, such as requirement of a large set of training data, robustness to imaging conditions not present in the training set, and complexity of the search process. In this paper we tackle the second problem by using a deep belief network learning architecture, and the third problem by resorting to efficient searching algorithms. As an example, we illustrate the performance of the algorithm in lip segmentation and tracking in video sequences. Quantitative comparison using different strategies for the search process are presented. We also compare our approach to a state-of-the-art segmentation and tracking algorithm. The comparison show that our algorithm produces competitive segmentation results and that efficient search strategies reduce ten times the run-complexity.|
|Keywords:||Deep belief networks; lip segmentation; optimization algorthims; search methods; tracking|
|Rights:||© 2010 IEEE|
|Appears in Collections:||Computer Science publications|
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.