Perceptual visual quality assessment in multimedia communication

Zhou, Wei and Amirpour, Hadi 2025. Perceptual visual quality assessment in multimedia communication. Presented at: MM '25: The 33rd ACM International Conference on Multimedia, Dublin, Ireland, 27-31 October 2025. MM '25: Proceedings of the 33rd ACM International Conference on Multimedia. ACM, pp. 14340-14341. 10.1145/3746027.3759205

Full text not available from this repository.

Abstract

The rapid expansion of multimedia services, such as video streaming, video conferencing, virtual reality, and cloud gaming, makes maintaining and evaluating high perceptual visual quality essential for user experience and system competitiveness. However, visual content can degrade at multiple stages, including acquisition, compression, transmission, enhancement, and display, where suboptimal enhancement may also introduce artifacts and reduce perceived quality. The core challenge is to reliably measure and predict this perceived quality so that it can be maintained or improved. Perceptual Visual Quality Assessment (PVQA) addresses this by evaluating visual quality from the perspective of human subjects, through subjective studies and objective prediction models. Beyond humans, recent work also extends PVQA to machines and robots, where the goal is to preserve downstream task performance (e.g., segmentation accuracy and planning success) under distortions or bandwidth constraints. This tutorial provides a concise, practice-oriented overview of PVQA: fundamentals and human vision considerations; image and video quality assessment; methods for immersive/3D media; opportunities and challenges in the era of foundation models and GenAI; perceptual optimization loops that close the gap between assessment and decisions in coding, streaming, and embodied perception; and domain applications. Finally, we summarize the key concepts, toolchains, and future opportunities for PVQA to be used in modern multimedia communication.

Item Type:	Conference or Workshop Item (Paper)
Date Type:	Published Online
Status:	Published
Schools:	Schools > Computer Science & Informatics
Publisher:	ACM
ISBN:	9798400720352
Last Modified:	18 Nov 2025 10:31
URI:	https://orca.cardiff.ac.uk/id/eprint/182485

Actions (repository staff only)

Edit Item