Segment-based free space estimation using plane normal vector in disparity space

Jeonghyun Seo, Changjae Oh, Kwanghoon Sohn

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

This paper proposes a framework of segment-based free space estimation using plane normal vector with stereo vision. An image is divided into compact superpixels and each of them is viewed as a plane composed of the normal vector in disparity space. To deal with the variation of illumination and shading in real traffic scenes, we estimate depth information for the segmented stereo pair. The representative normal vector is then computed at superpixel-level, which alleviates the problems of conventional color-based approaches and depth-based approaches simultaneously. Based on the assumption that the central-bottom of input image is navigable region, the free space is then determined by clustering the plane normal vectors with the K-means algorithm. In experiments, the proposed approach is evaluated on the KITTI dataset in which we provide the ground truth labels for free space region. The experimental results demonstrate that the proposed framework effectively estimates the free space under various real traffic scenes, and outperforms current state of the art methods both qualitatively and quantitatively.

Original languageEnglish
Title of host publication2015 International Conference on Connected Vehicles and Expo, ICCVE 2015 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages144-149
Number of pages6
ISBN (Electronic)9781509002641
DOIs
Publication statusPublished - 2016 Apr 4
EventInternational Conference on Connected Vehicles and Expo, ICCVE 2015 - Shenzhen, China
Duration: 2015 Oct 192015 Oct 23

Publication series

Name2015 International Conference on Connected Vehicles and Expo, ICCVE 2015 - Proceedings

Other

OtherInternational Conference on Connected Vehicles and Expo, ICCVE 2015
CountryChina
CityShenzhen
Period15/10/1915/10/23

All Science Journal Classification (ASJC) codes

  • Automotive Engineering
  • Computer Networks and Communications

Fingerprint Dive into the research topics of 'Segment-based free space estimation using plane normal vector in disparity space'. Together they form a unique fingerprint.

  • Cite this

    Seo, J., Oh, C., & Sohn, K. (2016). Segment-based free space estimation using plane normal vector in disparity space. In 2015 International Conference on Connected Vehicles and Expo, ICCVE 2015 - Proceedings (pp. 144-149). [07447584] (2015 International Conference on Connected Vehicles and Expo, ICCVE 2015 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICCVE.2015.6