AI-Assisted Energy-Efficient Multimedia Systems
The 17th ACM Multimedia System Conference (MMSys’26)
Hong Kong SAR
4th – 8th April 2026
Zoha Azimi
Abstract: Video streaming constitutes the majority of today’s Internet traffic and is expected to continue growing in scale, complexity, and environmental impact. As video systems, from encoding to delivery, consume substantial computational resources, their energy footprint has emerged as a critical challenge for both industry and research communities. At the same time, recent advances in Artificial Intelligence (AI) and Generative AI have improved video quality and adaptive streaming performance, yet often at the cost of increased computational load and higher energy consumption. This increase creates a growing need for streaming systems that not only deliver high Quality of Experience (QoE) but also minimize energy usage across heterogeneous devices and network infrastructures. The scope of this doctoral study is within the end-to-end video ecosystem, focusing on balancing the trade-off between energy consumption and user experience. It aims to investigate and develop intelligent frameworks that treat sustainability as a primary metric alongside performance, improving energy efficiency across the video lifecycle without compromising the viewer’s experience. We present three fundamental research questions to target the related challenges in the domain of sustainable multimedia systems.
