Authors:
Houda Maâmatou
1
;
Thierry Chateau
2
;
Sami Gazzah
3
;
Yann Goyat
4
and
Najoua Essoukri Ben Amara
3
Affiliations:
1
Blaise Pascal University, University of Sousse and Logiroad, France
;
2
Blaise Pascal University, France
;
3
University of Sousse, Tunisia
;
4
Logiroad, France
Keyword(s):
Transductive Transfer Learning, Specialization, Generic Classifier, Pedestrian Detection, Sequential Monte Carlo Filter (SMC).
Related
Ontology
Subjects/Areas/Topics:
Computer Vision, Visualization and Computer Graphics
;
Features Extraction
;
Image and Video Analysis
Abstract:
In this paper, we tackle the problem of domain adaptation to perform object-classification and detection tasks in
video surveillance starting by a generic trained detector. Precisely, we put forward a new transductive transfer
learning framework based on a sequential Monte Carlo filter to specialize a generic classifier towards a specific
scene. The proposed algorithm approximates iteratively the target distribution as a set of samples (selected
from both source and target domains) which feed the learning step of a specialized classifier. The output
classifier is applied to pedestrian detection into a traffic scene. We have demonstrated by many experiments,
on the CUHK Square Dataset and the MIT Traffic Dataset, that the performance of the specialized classifier
outperforms the generic classifier and that the suggested algorithm presents encouraging results.