Real-Time Table Plane Detection Using Accelerometer Information And Organized Point Cloud Data From Kinect Sensor

Le Van Hung, Vlaminck Michiel, Vu Hai, Nguyen Thi Thuy, Le Thi Lan, Tran Thi Thanh Hai, Luong Quang Hiep, Veelaert Peter, Wilfried Philips
Author affiliations

Authors

  • Le Van Hung
  • Vlaminck Michiel
  • Vu Hai
  • Nguyen Thi Thuy
  • Le Thi Lan
  • Tran Thi Thanh Hai
  • Luong Quang Hiep
  • Veelaert Peter
  • Wilfried Philips

DOI:

https://doi.org/10.15625/1813-9663/32/3/7689

Keywords:

Table plane detection, Accelerometer, Organized point cloud, Plane segmentation

Abstract

Table plane detection in the scene is a prerequisite step in nding and grasping theobject aids systems for visually impaired people. In order to determine table plane in the scene,rstly we have to detect planes in the scene and then dene the table plane from these detectedplanes based on the proper characteristic of the table. Even a number of approaches have beenproposed for plane segmentation, it still lacks the work for table plane detection. In this paper, wepropose a table plane detection method using information coming from Kinect. The contributionof our paper is three-fold. Firstly, for plane detection step, we apply down-sampling technique forplane segmentation using organized point cloud in order to get real time computation. Secondly,we propose to apply accelerometer information provided by Kinect to detect the table plane amongdetected planes. Finally, we dene three dierent measures for table plane detection evaluation. Theproposed method has been evaluated using a dataset of 10 scenes captured in the context of objectnding aid service for visually impaired people. The proposed method outperforms the state of theart method based on PROSAC and obtains a comparable result as a method based on organizedpoint cloud with the frame rate is 6 times higher.

Metrics

Metrics Loading ...

Downloads

Published

11-04-2017

How to Cite

[1]
L. V. Hung, “Real-Time Table Plane Detection Using Accelerometer Information And Organized Point Cloud Data From Kinect Sensor”, JCC, vol. 32, no. 3, p. 243–258, Apr. 2017.

Issue

Section

Articles