Szczegóły publikacji
Opis bibliograficzny
The accuracy of subjects in a quality experiment: a theoretical subject model / Lucjan JANOWSKI, Margaret Pinson // IEEE Transactions on Multimedia ; ISSN 1520-9210. — 2015 — vol. 17 no. 12, s. 2210–2224. — Bibliogr. s. 2224, Abstr. — Publikacja dostępna online od: 2015-10-02
Autorzy (2)
- AGHJanowski Lucjan
- Pinson Margaret H.
Słowa kluczowe
Dane bibliometryczne
ID BaDAP | 94788 |
---|---|
Data dodania do BaDAP | 2016-01-20 |
Tekst źródłowy | URL |
DOI | 10.1109/TMM.2015.2484963 |
Rok publikacji | 2015 |
Typ publikacji | artykuł w czasopiśmie |
Otwarty dostęp | |
Czasopismo/seria | IEEE Transactions on Multimedia |
Abstract
How accurately are pe ople able to use the absolute category rating (ACR) 5-level scale? Put another way, how repeatable are an individual subject’s scores? Several subjective experiments have asked subjects to rate the same sequences a couple of times. Anal yses indicate that none of the subjects exactly repeated their pr ior scores for these sequences. Wewouldliketobetterunderstandthisimperfection.Thispaper uses ACR subjective video quality tests to explore the precision of subjectiveratings. To makeformalmeasurementspossible,we propose a theoretical subject mo del that isthe maincontribution of this paper. The proposed subject model indicates three major factors that influence accuracy: su bject bias, subject inaccuracy, and stimulus scoring difficulty. These appear to be separate random effects and their existence is a reason why none of the subjectswereabletoperfectlyrepeatscores.Therearethreekey consequences. First, subject scoring behavior includes a random component that spans approximately half of the rating scale. Second, the sensitivity and accu racy of most subjective analyses canbeimprovedifthesubjectscoresarenormalizedbyremoving subject bias. Third, to some extent, multiple subjects can be replaced with a single subject who rates each sequence multiple times.