A Correlated Parts Model for Object Detection in Large 3D Scans

Simple item page
dc.contributor.authorSunkel, Martinen_US
dc.contributor.authorJansen, Silkeen_US
dc.contributor.authorWand, Michaelen_US
dc.contributor.authorSeidel, Hans-Peteren_US
dc.contributor.editorI. Navazo, P. Poulinen_US
dc.date.accessioned2015-02-28T15:22:37Z
dc.date.available2015-02-28T15:22:37Z
dc.date.issued2013en_US
dc.description.abstractThis paper addresses the problem of detecting objects in 3D scans according to object classes learned from sparse user annotation. We model objects belonging to a class by a set of fully correlated parts, encoding dependencies between local shapes of different parts as well as their relative spatial arrangement. For an efficient and comprehensive retrieval of instances belonging to a class of interest, we introduce a new approximate inference scheme and a corresponding planning procedure. We extend our technique to hierarchical composite structures, reducing training effort and modeling spatial relations between detected instances. We evaluate our method on a number of real-world 3D scans and demonstrate its benefits as well as the performance of the new inference algorithm.en_US
dc.description.seriesinformationComputer Graphics Forumen_US
dc.identifier.issn1467-8659en_US
dc.identifier.urihttps://doi.org/10.1111/cgf.12040en_US
dc.publisherThe Eurographics Association and Blackwell Publishing Ltd.en_US
dc.subjectComputer Graphics [I.3.5]en_US
dc.subjectComputational Geometry and Object Modelingen_US
dc.subjectObject hierarchiesen_US
dc.subjectImage Processing and Computer Vision [I.4.8]en_US
dc.subjectScene Analysisen_US
dc.subjectObject recognitionen_US
dc.subjectArtificial Intelligence [I.2.10]en_US
dc.subjectVision and Scene Understandingen_US
dc.subjectShapeen_US
dc.titleA Correlated Parts Model for Object Detection in Large 3D Scansen_US
Files
Collections
32-Issue 2
Full Papers 2013 - CGF 32-Issue 2