2026-09-19T09:07:50Zhttps://dspace.mit.edu/server/oai/requestoai:dspace.mit.edu:1721.1/854662026-06-06T00:48:55Zcom_1721.1_7582com_1721.1_7581col_1721.1_131023
Leslie Pack Kaelbling and Tomás Lozano-Pérez.
Popović, Sanja, M. Eng. Massachusetts Institute of Technology
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science.
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
2014-03-06T15:44:03Z
2014-03-06T15:44:03Z
2013
2013
http://hdl.handle.net/1721.1/85466
870969578
Thesis: M. Eng., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2013.
Cataloged from PDF version of thesis.
Includes bibliographical references (pages 40-41).
Modern graphics hardware (GPUs) are an amazing computational resource, but only for algorithms with suitable structure. Computer vision algorithms have many characteristics in common with computer graphics algorithms, in particular, they repeat some operations, such as feature computations, at many places in the image. However, there are also more global operations, such as finding nearest neighbors in feature space, that present more of a challenge. In this thesis, we showed how a state-of-the-art object detector, based on RGB-D images, could be parallelized for use on GPUs. By using nVidia's CUDA platform we improved the running times of critical sections up to 38 times. We also built a two-stage pipeline that improves multiple object detection in cluttered scenes. The first stage aims to achieve high precision, even at the cost of lower recall, by detecting only the less occluded objects. This results in large fraction of the scene being labeled which enables the algorithm in the second stage to focus on the less visible objects that would otherwise be missed. We analyze the performance of our algorithm and lay grounds for the future work and extensions.
by Sanja Popovic.
M.Eng.
41 pages
eng
Massachusetts Institute of Technology
M.I.T. theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission. See provided URL for inquiries about permission.
http://dspace.mit.edu/handle/1721.1/7582
Electrical Engineering and Computer Science.
Parallelized two-stage object detection in cluttered RGB-D scenes
Parallelized 2-stage object detection in cluttered RGB-D scenes
Thesis
application/pdf
false
Publication
unknown
unknown
<Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="1cb4d0d0-8b9a-4b2f-8787-dc65254d22c4">
<Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843</Type>
<Language>eng</Language>
<Title>Parallelized two-stage object detection in cluttered RGB-D scenes</Title>
<Subtitle>Parallelized 2-stage object detection in cluttered RGB-D scenes</Subtitle>
<PublishedIn>
<Publication>
</Publication>
</PublishedIn>
<PublicationDate>2013</PublicationDate>
<Authors>
<Author>
<DisplayName>Popović, Sanja, M. Eng. Massachusetts Institute of Technology</DisplayName>
<Affiliation>
<OrgUnit>
</OrgUnit>
</Affiliation>
</Author>
</Authors>
<Editors>
</Editors>
<Publishers>
<Publisher>
<DisplayName>Massachusetts Institute of Technology</DisplayName>
<OrgUnit />
</Publisher>
</Publishers>
<License>http://dspace.mit.edu/handle/1721.1/7582</License>
<Keyword>Electrical Engineering and Computer Science.</Keyword>
<Abstract>Modern graphics hardware (GPUs) are an amazing computational resource, but only for algorithms with suitable structure. Computer vision algorithms have many characteristics in common with computer graphics algorithms, in particular, they repeat some operations, such as feature computations, at many places in the image. However, there are also more global operations, such as finding nearest neighbors in feature space, that present more of a challenge. In this thesis, we showed how a state-of-the-art object detector, based on RGB-D images, could be parallelized for use on GPUs. By using nVidia's CUDA platform we improved the running times of critical sections up to 38 times. We also built a two-stage pipeline that improves multiple object detection in cluttered scenes. The first stage aims to achieve high precision, even at the cost of lower recall, by detecting only the less occluded objects. This results in large fraction of the scene being labeled which enables the algorithm in the second stage to focus on the less visible objects that would otherwise be missed. We analyze the performance of our algorithm and lay grounds for the future work and extensions.</Abstract>
<Access xmlns="http://purl.org/coar/access_right"
>
</Access>
</Publication>