Institute for Telecommunication Sciences / Research Topics / Video Quality Research / Software / INLSA

An iterated nested least-squares algorithm (INLSA) for fitting multiple data sets

INLSA is an algorithm that allows multiple subjective datasets to be fitted to a single subjective scale. INLSA computes the fit from a common set of objective metrics. 

Click here for a reference. This document fully describes the algorithm.

Click here for a comparison of INLSA and subjective mapping. This paper demonstrates why data from different subjective tests must be fitted.

Click here for an example of how multiple subjective datasets can be combined using overlapping subjective datasets.


Click here for MATLAB® code. This code may be used for any purpose, commercial or non-commercial. Please contact Margaret Pinson if you find any bugs or errors in this code.

  • Files inlsa.m and pars_inlsa.m implement INLSA.
  • File inlsa_demo.m creates made-up data for three (3) experiments and plots that data before and after running INLSA. The user can actually see what INLSA does to the data. Also the user gets a concrete example of how to call INLSA.  The user can just replace the made-up data with real data and the use INLSA.