HDR Display Quality Evaluation by incorporating Perceptual Component Models into a Machine Learning framework

作者：

Highlights：

• A perceptual model based on visual system is better at describing display quality than a physical model.

• Consolidating predictions from individual parameters is better than using one model with multiple parameters.

• Machine learning techniques (ML) are better at prediction than linear regression.

• Combining ML methods with perceptual model improve prediction when tested on unknown conditions.

摘要

•A perceptual model based on visual system is better at describing display quality than a physical model.•Consolidating predictions from individual parameters is better than using one model with multiple parameters.•Machine learning techniques (ML) are better at prediction than linear regression.•Combining ML methods with perceptual model improve prediction when tested on unknown conditions.

论文关键词：Display quality assessment,High Dynamic Range (HDR),Subjective study,Machine learning,Visual quality

论文评审过程：Received 24 September 2018, Revised 14 February 2019, Accepted 18 February 2019, Available online 5 March 2019, Version of Record 12 March 2019.

论文官网地址：https://doi.org/10.1016/j.image.2019.02.007