Object segmentation and tracking from sequential stereo images for 3D object modeling

Title
Object segmentation and tracking from sequential stereo images for 3D object modeling
Authors
한인규김형년김경구박지형
Keywords
3D object modeling; stereo image; object segmentation; object tracking
Issue Date
2010-01
Publisher
HCI KOREA 2010
Abstract
The previous researches of 3D object modeling were performed in a limited area that has no other objects or background. However, in order to apply the object modeling to the real environment where various objects and background are mixed with, the process that segments the target object from environment is needed. In this paper, we propose a method of segmenting the object region from stereo images and tracking the object region through sequential images to build a 3D object model. The object region can be segmented more robustly using depth information than using color information, and it can be tracked on the sequential images by referring the distribution of the target area because the images have small changes in viewpoints. Experimental results show that our method can robustly segment and track the object region in the real environment, and show that the execution time of the object modeling is improved.
URI
http://pubs.kist.re.kr/handle/201004/37178
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