Vision-based place categorization

dc.contributor.advisor Christensen, Henrik I.
dc.contributor.author Bormann, Richard Klaus Eduard en_US
dc.contributor.committeeMember Ahamad, Mustaque
dc.contributor.committeeMember Copeland, John
dc.contributor.committeeMember Feamster, Nicholas G.
dc.contributor.committeeMember Traynor, Patrick
dc.contributor.department Computing en_US
dc.date.accessioned 2011-03-04T20:59:30Z
dc.date.available 2011-03-04T20:59:30Z
dc.date.issued 2010-11-18 en_US
dc.description.abstract In this thesis we investigate visual place categorization by combining successful global image descriptors with a method of visual attention in order to automatically detect meaningful objects for places. The idea behind this is to incorporate information about typical objects for place categorization without the need for tedious labelling of important objects. Instead, the applied attention mechanism is intended to find the objects a human observer would focus first, so that the algorithm can use their discriminative power to conclude the place category. Besides this object-based place categorization approach we employ the Gist and the Centrist descriptor as holistic image descriptors. To access the power of all these descriptors we employ SVM-DAS (discriminative accumulation scheme) for cue integration and furthermore smooth the output trajectory with a delayed Hidden Markov Model. For the classification of the variety of descriptors we present and evaluate several classification methods. Among them is a joint probability modelling approach with two approximations as well as a modified KNN classifier, AdaBoost and SVM. The latter two classifiers are enhanced for multi-class use with a probabilistic computation scheme which treats the individual classifiers as peers and not as a hierarchical sequence. We check and tweak the different descriptors and classifiers in extensive tests mainly with a dataset of six homes. After these experiments we extend the basic algorithm with further filtering and tracking methods and evaluate their influence on the performance. Finally, we also test our algorithm within a university environment and on a real robot within a home environment. en_US
dc.description.degree M.S. en_US
dc.identifier.uri http://hdl.handle.net/1853/37233
dc.publisher Georgia Institute of Technology en_US
dc.subject Vision-based place categorization en_US
dc.subject Multi-class SVM en_US
dc.subject Multi-class AdaBoost en_US
dc.subject.lcsh Algorithms
dc.subject.lcsh Robot vision
dc.subject.lcsh Robotics
dc.subject.lcsh Robotics Human factors
dc.title Vision-based place categorization en_US
dc.type Text
dc.type.genre Thesis
dspace.entity.type Publication
local.contributor.advisor Christensen, Henrik I.
local.contributor.corporatename College of Computing
local.contributor.corporatename Institute for Robotics and Intelligent Machines (IRIM)
local.relation.ispartofseries Computer Science Graduate Program
relation.isAdvisorOfPublication afdc727f-2705-4744-945f-e7d414f2212b
relation.isOrgUnitOfPublication c8892b3c-8db6-4b7b-a33a-1b67f7db2021
relation.isOrgUnitOfPublication 66259949-abfd-45c2-9dcc-5a6f2c013bcf
relation.isSeriesOfPublication 41e6384f-fa8d-4c63-917f-a26900b10f64
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
13.28 MB
Adobe Portable Document Format