Skip to main navigation Skip to search Skip to main content

Markerless Visual Servoing Using Virtual Face Tag and Image Moment Invariants

  • Arif Zubair
  • , Yili Fu
  • , Amjad Sohail
  • , Anwar Khadija
  • Harbin Institute of Technology
  • Divisional Head TandD Hr Directorate General
  • NED University of Engineering and Technology

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

Abstract

A novel normalized-centered image moment invariant-based visual servoing scheme is presented. A fresh combination of the moments invariants is proposed for planar symmetrical objects that are parallel to the image plane at the desired position. The output of the face detection module in the image plane, that is a rectangle around the face is used as a virtual target, treated as a planar symmetrical object which always remains parallel to the image plane. The proposed method completely decouples the DoF and is 2D translation, rotation, and scaling invariant. It was experimentally validated on a real robotic arm using a haar cascade-based face detection system to accomplish visual servoing tasks for both 2D printed faces and actual 3D faces. Unlike traditional deep learning based visual servoing schemes, the proposed markerless visual servoing technique does not require a 3D model of the face or a reference image/pose at the desired position and it can also work with unfamiliar faces without prior training.

Original languageEnglish
Title of host publication2022 4th International Conference on Robotics and Computer Vision, ICRCV 2022
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages123-128
Number of pages6
ISBN (Electronic)9781665481700
DOIs
StatePublished - 2022
Event4th International Conference on Robotics and Computer Vision, ICRCV 2022 - Virtual, Online, China
Duration: 25 Sep 202227 Sep 2022

Publication series

Name2022 4th International Conference on Robotics and Computer Vision, ICRCV 2022

Conference

Conference4th International Conference on Robotics and Computer Vision, ICRCV 2022
Country/TerritoryChina
CityVirtual, Online
Period25/09/2227/09/22

Keywords

  • face detection
  • IBVS
  • Image moment
  • markerless visual servoing
  • moment invariant

Fingerprint

Dive into the research topics of 'Markerless Visual Servoing Using Virtual Face Tag and Image Moment Invariants'. Together they form a unique fingerprint.

Cite this