Object detection with large intra-class variation

MOspace/Manakin Repository

Breadcrumbs Navigation

Object detection with large intra-class variation

Please use this identifier to cite or link to this item: http://hdl.handle.net/10355/14534

[-] show simple item record

dc.contributor.advisor Han, Tony X. en_US
dc.contributor.author Chen, Guang
dc.date.accessioned 2012-06-06T17:26:12Z
dc.date.available 2012-06-06T17:26:12Z
dc.date.issued 2011
dc.date.submitted 2011 Fall en_US
dc.identifier.other ChenG-121611-T493
dc.identifier.uri http://hdl.handle.net/10355/14534
dc.description Title from PDF of title page (University of Missouri--Columbia, viewed on June 6, 2012). en_US
dc.description The entire thesis text is included in the research.pf file; the official abstract appears in the short.pf file; a non-technical public abstract appears in the public.pf file. en_US
dc.description Thesis advisor: Dr. Tony X. Han en_US
dc.description Includes bibliographical references. en_US
dc.description M.S. University of Missouri-Columbia 2011. en_US
dc.description Dissertations, Academic -- University of Missouri--Columbia -- Electrical engineering. en_US
dc.description "December 2011" en_US
dc.description.abstract For object detection, the state-of-the-art performance is achieved through supervised learning. The performances of object detectors of this kind are mainly determined by two factors: features and underlying classification algorithms. In this work, we aim at improving the performance of object detectors from the aspect of classification algorithm. Observing the fact that classifiers used for object detection are task dependent and data driven, we developed a hybrid learning algorithm combining global classification and local adaptations, which automatically adjusts model complexity according to data distribution. We divide data samples into two groups, easy samples and ambiguous samples, using a learned global classifier. A local adaptation approach based on spectral clustering and proposed Min-Max model adaptation is then applied to further process the ambiguous samples. The proposed algorithm automatically determines model complexity of the local learning algorithm according to the distribution of ambiguous samples. By autonomously striking a balance between model complexity and learning capacity, the proposed hybrid learning algorithm incarnates a human detector outperforming the state-of-the-art algorithms on a couple of benchmark datasets and a self-collected pedestrian dataset. Besides, the proposed Min-Max model adaptation algorithm also successfully improve the performance of an offline-trained classifier on-site by adapting the classifier towards newly acquired data, without worries about the tuning the adaptation rate parameter, which affects the performance gain substantially. Taking the object detection as a testbed, we implement an adapted object detector based on binary classification. Under different adaptation scenarios and different datasets including PASCAL, ImageNet, INRIA, and TUD Pedestrian, the proposed adaption method achieves significant performance gain and is compared favorably with the state-of-the-art adaptation method with the fine tuned adaptation rate. en_US
dc.format.extent viii, 46 pages en_US
dc.language.iso en_US en_US
dc.publisher University of Missouri--Columbia en_US
dc.relation.ispartof 2011 Freely available theses (MU) en_US
dc.subject object detection en_US
dc.subject classifier cascade en_US
dc.subject local learning en_US
dc.subject hybrid learning en_US
dc.title Object detection with large intra-class variation en_US
dc.type Thesis en_US
thesis.degree.discipline Electrical engineering en_US
thesis.degree.grantor University of Missouri--Columbia en_US
thesis.degree.name M.S. en_US
thesis.degree.level Masters en_US
dc.relation.ispartofcommunity University of Missouri-Columbia. Graduate School. Theses and Dissertations. Theses. 2011 Theses


This item appears in the following Collection(s)

[-] show simple item record