2018/07/02 by Haiqiang Wang, Wang, Haiqiang, Xinfeng Zhang +5
Computer Science · #Advanced Image Processing Techniques #FOS: Computer and information sciences #Image Enhancement Techniques #Image and Video Quality Assessment #Multimedia (cs.MM)
paper · pdf · doi:10.48550/arxiv.1807.00920
openalex publication_date 2018/07/02 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
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.