Integrating Depth Information in Shape Alignment of 3D Deformable Objects

Title
Integrating Depth Information in Shape Alignment of 3D Deformable Objects
Authors
김준식오상록
Keywords
Depth image; Landmark observation; Shape alignment; Deformable object
Issue Date
2014-02
Publisher
20th Korea-Japan Joint Workshop on Frontiers of Computer Vision
Citation
, 54-59
Abstract
We present a method to integrate depth information in aligning a shape of deformable object on the observed data from an RGBD sensor. We extend the constrained local model which is widely used in the shape alignment on an image, by augmenting the landmark observation probability from depth image patches around the landmark locations. Experiments show that the simple extension integrating depth information greatly improves the localization of landmark, and we discuss about the practical considerations in the depth data fusion as well.
URI
http://pubs.kist.re.kr/handle/201004/47225
Appears in Collections:
KIST Publication > Conference Paper
Files in This Item:
There are no files associated with this item.
Export
RIS (EndNote)
XLS (Excel)
XML


qrcode

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

BROWSE