A JND-based Video Quality Assessment Model and Its Application
This work addresses video quality assessment for applications in media and telecommunications, but it is incremental as it builds on existing JND-based datasets and methods.
The authors tackled the problem of video quality assessment by proposing a model based on the Just-Noticeable-Difference criterion that accounts for both subject and content variabilities, using a probabilistic framework to clean subjective test data, and demonstrated its effectiveness in filtering unreliable scores from the VideoSet dataset.
Based on the Just-Noticeable-Difference (JND) criterion, a subjective video quality assessment (VQA) dataset, called the VideoSet, was constructed recently. In this work, we propose a JND-based VQA model using a probabilistic framework to analyze and clean collected subjective test data. While most traditional VQA models focus on content variability, our proposed VQA model takes both subject and content variabilities into account. The model parameters used to describe subject and content variabilities are jointly optimized by solving a maximum likelihood estimation (MLE) problem. As an application, the new subjective VQA model is used to filter out unreliable video quality scores collected in the VideoSet. Experiments are conducted to demonstrate the effectiveness of the proposed model.