Quality assessment aims to measure the degree of delight or annoyance of the users when experiencing an application or service. With the quick improvement of content acquisition, processing, transmission, and display techniques, the end-users are expecting and demanding continuously improved quality of experience from the service providers. To guarantee a good quality of experience to end-users, perceptual quality assessment is introduced and widely studied in recent years. Since the ultimate receiver of the processed signal is usually human, it is reasonable and beneficial to take human perception properties into consideration. Though we still have limited knowledge of the intrinsic neuroscience working mechanism of human perception, it is worthwhile to study and take inspiration from neuroscience and utilize these properties for computational modeling of perceptual quality.
Many of the current quality assessment models have already attempted to include human perception properties at some level, however, the majority of these models only take simplified concepts of human perception, and use ‘black box’ machine learning techniques to model the quality of experience. The rapid development of neuroscience and computer science have provided opportunities for deeper explorations of the intrinsic neuroscience working mechanism of quality perception, and to utilize computational neuroscience theories and models for more efficient and explainable quality assessment. Specifically, on one hand the underlying biological bases of human perception especially those related to quality perception can be further explored on the basis of the recent advancement of neurobiology. While on the other hand, it is worthwhile to seek better ways to apply the relevant neuroscience working mechanisms for quality assessment and to build more accurate brain-inspired computational quality assessment models.
This Research Topic solicits novel and high-quality papers to present computational neuroscience studies for perceptual quality assessment and the potential applications in artificial systems. The topics include, but are not limited to:
1) Neuroscience studies of human perception, especially those related to quality perception;
2) Computational neuroscience models for perceptual quality modeling;
3) Neuroscience inspired perceptual quality modeling, including perceptual quality assessment, control, and optimization;
4) Neuroscience inspired visual attention modeling, including the mechanism of visual attention, visual saliency prediction and the utilization of visual attention models in relevant applications;
5) Quality assessment based on advanced learning technologies, such as deep learning, transfer learning, unsupervised learning, contrastive learning, etc.
6) Quality assessment for emerging multimedia technologies, such as virtual reality, augmented reality, light fields, high dynamic range media, etc.
Quality assessment aims to measure the degree of delight or annoyance of the users when experiencing an application or service. With the quick improvement of content acquisition, processing, transmission, and display techniques, the end-users are expecting and demanding continuously improved quality of experience from the service providers. To guarantee a good quality of experience to end-users, perceptual quality assessment is introduced and widely studied in recent years. Since the ultimate receiver of the processed signal is usually human, it is reasonable and beneficial to take human perception properties into consideration. Though we still have limited knowledge of the intrinsic neuroscience working mechanism of human perception, it is worthwhile to study and take inspiration from neuroscience and utilize these properties for computational modeling of perceptual quality.
Many of the current quality assessment models have already attempted to include human perception properties at some level, however, the majority of these models only take simplified concepts of human perception, and use ‘black box’ machine learning techniques to model the quality of experience. The rapid development of neuroscience and computer science have provided opportunities for deeper explorations of the intrinsic neuroscience working mechanism of quality perception, and to utilize computational neuroscience theories and models for more efficient and explainable quality assessment. Specifically, on one hand the underlying biological bases of human perception especially those related to quality perception can be further explored on the basis of the recent advancement of neurobiology. While on the other hand, it is worthwhile to seek better ways to apply the relevant neuroscience working mechanisms for quality assessment and to build more accurate brain-inspired computational quality assessment models.
This Research Topic solicits novel and high-quality papers to present computational neuroscience studies for perceptual quality assessment and the potential applications in artificial systems. The topics include, but are not limited to:
1) Neuroscience studies of human perception, especially those related to quality perception;
2) Computational neuroscience models for perceptual quality modeling;
3) Neuroscience inspired perceptual quality modeling, including perceptual quality assessment, control, and optimization;
4) Neuroscience inspired visual attention modeling, including the mechanism of visual attention, visual saliency prediction and the utilization of visual attention models in relevant applications;
5) Quality assessment based on advanced learning technologies, such as deep learning, transfer learning, unsupervised learning, contrastive learning, etc.
6) Quality assessment for emerging multimedia technologies, such as virtual reality, augmented reality, light fields, high dynamic range media, etc.