mirage   mirage   mirage

From multitarget tracking to event recognition in videos

DSpace/Manakin Repository

Show simple item record

dc.contributor.advisor Todorovic, Sinisa
dc.creator Brendel, William
dc.date.accessioned 2011-05-12T22:13:29Z
dc.date.available 2013-10-15T22:18:26Z
dc.date.copyright 2011-04-25
dc.date.issued 2011-05-12
dc.identifier.uri http://hdl.handle.net/1957/21315
dc.description Graduation date: 2011 en_US
dc.description Access restricted to the OSU Community at author's request from May 12, 2011 - May 12, 2012
dc.description.abstract This dissertation addresses two fundamental problems in computer vision—namely, multitarget tracking and event recognition in videos. These problems are challenging because uncertainty may arise from a host of sources, including motion blur, occlusions, and dynamic cluttered backgrounds. We show that these challenges can be successfully addressed by using a multiscale, volumetric video representation, and taking into account various constraints between events offered by domain knowledge. The dissertation presents our two alternative approaches to multitarget tracking. The first approach seeks to transitively link object detections across consecutive video frames by finding the maximum independent set of a graph of all object detections. Two maximum-independent-set algorithms are specified, and their convergence properties theoretically analyzed. The second approach hierarchically partitions the space-time volume of a video into tracks of objects, producing a segmentation graph of that video. The resulting tracks encode rich contextual cues between salient video parts in space and time, and thus facilitate event recognition, and segmentation in space and time. We also describe our two alternative approaches to event recognition. The first approach seeks to learn a structural probabilistic model of an event class from training videos represented by hierarchical segmentation graphs. The graph model is then used for inference of event occurrences in new videos. Learning and inference algorithms are formulated within the same framework, and their convergence rates theoretically analyzed. The second approach to event recognition uses probabilistic first-order logic for reasoning over continuous time intervals. We specify the syntax, learning, and inference algorithms of this probabilistic event logic. Qualitative and quantitative results on benchmark video datasets are also presented. The results demonstrate that our approaches provide consistent video interpretation with respect to acquired domain knowledge. We outperform most of the state-of-the-art approaches on benchmark datasets. We also present our new basketball dataset that complements existing benchmarks with new challenges. en_US
dc.language.iso en_US en_US
dc.subject multitarget tracking en_US
dc.subject maximum weighted independent set en_US
dc.subject graph learning en_US
dc.subject video segmentation en_US
dc.subject event recognition en_US
dc.subject.lcsh Computer vision en_US
dc.subject.lcsh Pattern recognition systems
dc.title From multitarget tracking to event recognition in videos en_US
dc.type Thesis/Dissertation en_US
dc.degree.name Doctor of Philosophy (Ph. D.) in Computer Science en_US
dc.degree.level Doctoral en_US
dc.degree.discipline Engineering en_US
dc.degree.grantor Oregon State University en_US
dc.contributor.committeemember Dietterich, Thomas
dc.contributor.committeemember Fern, Alan
dc.contributor.committeemember Raich, Raviv
dc.description.embargopolicy OSU Users


This item appears in the following Collection(s)

Show simple item record

Search ScholarsArchive@OSU


Advanced Search

Browse

My Account

Statistics