A matlab implementation of Bradley–Terry (BT) model [1]. BT model is the most popular approache to convert winning frequency matrix obtained from pairwise comparison (PC) experiment to quality scores (or BT scores). Typically, in PC experiments, after outlier detection, the subjective scores from the observers are represented in the wining frequency matrix and then translated to absolute quality scores using the Bradley-Terry (BT) model.
- Matlab with version > R2015
Check the Main.m for some simple tests
- [1] R. A. Bradley, Paired Comparisons: Some Basic Procedures and Examples. Elsevier Science Publishers, Feb. 1984.
- [2] J. C. Handley, “Comparative Analysis of Bradley-Terry and Thurstone-Mosteller Paired Comparison Models for Image Quality Assessment,” in PICS : Image: Processing, Quality, Capture, Systems, Montreal, Canada, Apr. 2001.
- [3] F. Jabar, J. Ascenso, and M.P. Queluz, “Field-of-View Effect on the Perceived Quality of Omnidirectional Images,” Proc. of the IEEE International Conference on Multimedia & Expo Workshops (ICMEW), Athlone, Ireland, Jul. 2020.