BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Psychometric Scaling of TID2013 Dataset - Aliaksei Mikhailiuk (Uni
 versity of Cambridge)
DTSTART:20180517T131500Z
DTEND:20180517T140000Z
UID:TALK105883@talks.cam.ac.uk
CONTACT:Alexander Simpson
DESCRIPTION:TID2013 is a subjective image quality assessment dataset with 
 a wide range of distortion types and over 3000 images. The dataset has pro
 ven to be a challenging test for objective quality metrics. The dataset me
 an opinion scores were obtained by collecting pairwise comparison judgment
 s using the Swiss tournament system\, and averaging votes of observers. Ho
 wever\, this approach differs from the usual analysis of multiple pairwise
  comparisons\, which involves psychometric scaling of the comparison data 
 using either Thurstone or Bradley-Terry models. In this paper we investiga
 te how quality scores change when they are computed using such psychometri
 c scaling instead of averaging vote counts. In order to properly scale TID
 2013 quality scores\, we conduct four additional experiments of two differ
 ent types\, which we found necessary to produce a common quality scale: co
 mparisons with reference images\, and cross-content comparisons. We demons
 trate on a fifth validation experiment that the two additional types of co
 mparisons are necessary and in conjunction with psychometric scaling impro
 ve the consistency of quality scores\, especially across images depicting 
 different contents.
LOCATION:SS03 Meeting Room\, Computer Laboratory
END:VEVENT
END:VCALENDAR