Expertise screening in crowdsourcing image quality

Vlad Hosu, Hanhe Lin, Dietmar Saupe

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

11 Citations (Scopus)

Abstract

We propose a screening approach to find reliable and effectively expert crowd workers in image quality assessment (IQA). Our method measures the users' ability to identify image degradations by using test questions, together with several relaxed reliability checks. We conduct multiple experiments, obtaining reproducible results with a high agreement between the expertise-screened crowd and the freelance experts of 0.95 Spearman rank order correlation (SROCC), with one restriction on the image type. Our contributions include a reliability screening method for uninformative users, a new type of test questions that rely on our proposed database1of pristine and artificially distorted images, a group agreement extrapolation method and an analysis of the crowdsourcing experiments.

Original languageEnglish
Title of host publication2018 10th International Conference on Quality of Multimedia Experience, QoMEX 2018
PublisherIEEE
Number of pages6
ISBN (Electronic)978-1-5386-2605-4
ISBN (Print)978-1-5386-2606-1
DOIs
Publication statusPublished - 13 Sept 2018
Event10th International Conference on Quality of Multimedia Experience, QoMEX 2018 - Sardinia, Italy
Duration: 29 May 20181 Jun 2018

Publication series

Name2018 10th International Conference on Quality of Multimedia Experience, QoMEX 2018
PublisherIEEE
ISSN (Electronic)2472-7814

Conference

Conference10th International Conference on Quality of Multimedia Experience, QoMEX 2018
Country/TerritoryItaly
CitySardinia
Period29/05/181/06/18

Keywords

  • crowdsourcing
  • expertise screening
  • image quality assessment
  • reliability screening

ASJC Scopus subject areas

  • Media Technology
  • Safety, Risk, Reliability and Quality

Fingerprint

Dive into the research topics of 'Expertise screening in crowdsourcing image quality'. Together they form a unique fingerprint.

Cite this