Transductive Transfer Learning to Specialize a Generic Classifier Towards a Specific Scene

Houda Maâmatou, Thierry Chateau, Sami Gazzah, Yann Goyat, Najoua Essoukri Ben Amara

2016

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.

Paper Citation

in Harvard Style

Maâmatou H., Chateau T., Gazzah S., Goyat Y. and Essoukri Ben Amara N. (2016). Transductive Transfer Learning to Specialize a Generic Classifier Towards a Specific Scene . In Proceedings of the 11th Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP, (VISIGRAPP 2016) ISBN 978-989-758-175-5, pages 411-422. DOI: 10.5220/0005725104110422

in Bibtex Style

@conference{visapp16,
author={Houda Maâmatou and Thierry Chateau and Sami Gazzah and Yann Goyat and Najoua Essoukri Ben Amara},
title={Transductive Transfer Learning to Specialize a Generic Classifier Towards a Specific Scene},
booktitle={Proceedings of the 11th Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP, (VISIGRAPP 2016)},
year={2016},
pages={411-422},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0005725104110422},
isbn={978-989-758-175-5},
}

in EndNote Style

TY - CONF
JO - Proceedings of the 11th Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP, (VISIGRAPP 2016)
TI - Transductive Transfer Learning to Specialize a Generic Classifier Towards a Specific Scene
SN - 978-989-758-175-5
AU - Maâmatou H.
AU - Chateau T.
AU - Gazzah S.
AU - Goyat Y.
AU - Essoukri Ben Amara N.
PY - 2016
SP - 411
EP - 422
DO - 10.5220/0005725104110422