Segmentation of urban traffic scene based on 3D structure

Author

Zheng, Tan Lun ; Li-min, Xia ; Yanfei, Liu

Author_Institution

Coll. of Inf. Sci. & Eng., Central South Univ., Changsha, China

fYear

2011

fDate

21-23 Oct. 2011

Firstpage

240

Lastpage

243

Abstract

This paper proposes an approach for image segmentation of the urban traffic scene captured from a car-mounted camera. First of all, an improved SIFT feature matching algorithm is adopted for extracting 2D keypoints of the scene. Tracking the 2D keypoints generates the 3D points clouds that can estimate the 3D world structure and motion features. And then Multiple Kernels Support Vector Machines (MKSVM) is employed for sematic segmentation based on motion-derived 3D structure and SIFT features. Experiments show the efficiency and the relevancy of our approach.

Keywords

cameras; feature extraction; image matching; image motion analysis; image segmentation; support vector machines; traffic engineering computing; 2D keypoints extraction; 3D points clouds; 3D structure; SIFT feature matching; car-mounted camera; image segmentation; motion features; multiple kernels support vector machines; urban traffic scene; Cameras; Image segmentation; Kernel; Motion segmentation; Roads; Support vector machines; Three dimensional displays; 3D structure; MKSVM; SIFT; image segmentation; urban traffic;

fLanguage

English

Publisher

ieee

Conference_Titel

Image Analysis and Signal Processing (IASP), 2011 International Conference on

Conference_Location

Hubei

Print_ISBN

978-1-61284-879-2

Type

conf

DOI

10.1109/IASP.2011.6109038

Filename

6109038