Self-Occlusions and Disocclusions in Causal Video Object Segmentation

Handle URI:
http://hdl.handle.net/10754/621290
Title:
Self-Occlusions and Disocclusions in Causal Video Object Segmentation
Authors:
Yang, Yanchao; Sundaramoorthi, Ganesh ( 0000-0003-3471-6384 ) ; Soatto, Stefano
Abstract:
We propose a method to detect disocclusion in video sequences of three-dimensional scenes and to partition the disoccluded regions into objects, defined by coherent deformation corresponding to surfaces in the scene. Our method infers deformation fields that are piecewise smooth by construction without the need for an explicit regularizer and the associated choice of weight. It then partitions the disoccluded region and groups its components with objects by leveraging on the complementarity of motion and appearance cues: Where appearance changes within an object, motion can usually be reliably inferred and used for grouping. Where appearance is close to constant, it can be used for grouping directly. We integrate both cues in an energy minimization framework, incorporate prior assumptions explicitly into the energy, and propose a numerical scheme. © 2015 IEEE.
KAUST Department:
King Abdullah University of Science and Technology (KAUST), Saudi Arabia
Citation:
Yang Y, Sundaramoorthi G, Soatto S (2015) Self-Occlusions and Disocclusions in Causal Video Object Segmentation. 2015 IEEE International Conference on Computer Vision (ICCV). Available: http://dx.doi.org/10.1109/ICCV.2015.501.
Publisher:
Institute of Electrical and Electronics Engineers (IEEE)
Journal:
2015 IEEE International Conference on Computer Vision (ICCV)
Conference/Event name:
15th IEEE International Conference on Computer Vision, ICCV 2015
Issue Date:
19-Feb-2016
DOI:
10.1109/ICCV.2015.501
Type:
Conference Paper
Appears in Collections:
Conference Papers

Full metadata record

DC FieldValue Language
dc.contributor.authorYang, Yanchaoen
dc.contributor.authorSundaramoorthi, Ganeshen
dc.contributor.authorSoatto, Stefanoen
dc.date.accessioned2016-11-03T06:56:58Z-
dc.date.available2016-11-03T06:56:58Z-
dc.date.issued2016-02-19en
dc.identifier.citationYang Y, Sundaramoorthi G, Soatto S (2015) Self-Occlusions and Disocclusions in Causal Video Object Segmentation. 2015 IEEE International Conference on Computer Vision (ICCV). Available: http://dx.doi.org/10.1109/ICCV.2015.501.en
dc.identifier.doi10.1109/ICCV.2015.501en
dc.identifier.urihttp://hdl.handle.net/10754/621290-
dc.description.abstractWe propose a method to detect disocclusion in video sequences of three-dimensional scenes and to partition the disoccluded regions into objects, defined by coherent deformation corresponding to surfaces in the scene. Our method infers deformation fields that are piecewise smooth by construction without the need for an explicit regularizer and the associated choice of weight. It then partitions the disoccluded region and groups its components with objects by leveraging on the complementarity of motion and appearance cues: Where appearance changes within an object, motion can usually be reliably inferred and used for grouping. Where appearance is close to constant, it can be used for grouping directly. We integrate both cues in an energy minimization framework, incorporate prior assumptions explicitly into the energy, and propose a numerical scheme. © 2015 IEEE.en
dc.publisherInstitute of Electrical and Electronics Engineers (IEEE)en
dc.titleSelf-Occlusions and Disocclusions in Causal Video Object Segmentationen
dc.typeConference Paperen
dc.contributor.departmentKing Abdullah University of Science and Technology (KAUST), Saudi Arabiaen
dc.identifier.journal2015 IEEE International Conference on Computer Vision (ICCV)en
dc.conference.date11 December 2015 through 18 December 2015en
dc.conference.name15th IEEE International Conference on Computer Vision, ICCV 2015en
dc.contributor.institutionUniversity of California, Los Angeles, United Statesen
kaust.authorSundaramoorthi, Ganeshen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.