A two-step pedestrian detection algorithm based on RGB-D data

Qiming Li, Liqing Hu, Yaping Gao, Yimin Chen, Lizhuang Ma

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

A two-step pedestrian detection algorithm based RGB-D data is presented. Firstly, down-sample the depth image and extract the key information with a voxel grid. Second, remove the ground by using the random sample consensus (RANSAC) segmentation algorithm. Third, describe pedestrian characteristics by using Point Feature Histogram (PFH) and estimate the position of pedestrian preliminarily. Finally, calculate the pedestrian characteristics based on color image by Histogram of Oriented Gradient (HOG) descriptor and detect pedestrian using Support Vector Machine (SVM) classifier. The experimental results show that the algorithm can accurately detect the pedestrian not only in the single-pedestrian scene with pose variety but also in the multi-pedestrian scene with partial occlusion between pedestrians.

Original languageEnglish
Title of host publicationIntelligence Science I - 2nd IFIP TC 12 International Conference, ICIS 2017, Proceedings
EditorsZhongzhi Shi, Ben Goertzel, Jiali Feng
PublisherSpringer New York LLC
Pages288-294
Number of pages7
ISBN (Print)9783319681207
DOIs
StatePublished - 2017
Externally publishedYes
Event2nd IFIP TC 12 International Conference on Intelligence Science, ICIS 2017 - Shanghai, China
Duration: 25 Oct 201728 Oct 2017

Publication series

NameIFIP Advances in Information and Communication Technology
Volume510
ISSN (Print)1868-4238

Conference

Conference2nd IFIP TC 12 International Conference on Intelligence Science, ICIS 2017
Country/TerritoryChina
CityShanghai
Period25/10/1728/10/17

Keywords

  • Down-sample
  • HOG
  • PFH
  • Pedestrian detection
  • RGB-D
  • SVM

Fingerprint

Dive into the research topics of 'A two-step pedestrian detection algorithm based on RGB-D data'. Together they form a unique fingerprint.

Cite this