Cygnus AI Application
A Python library and example application that lets you apply AI algorithms to live FFmpeg streams and broadcast the processed output via RTMP. It also provides optional integrations for alerts, GPS telemetry via Kafka, and snapshot storage in MinIO.
https://github.com/annavalentina/CygnusAI
T. Anastasiou and I. Pastellas, “Cygnus dataset addressing civilian application challenges,”
2025. Available at: https://zenodo.org/records/14864373
[1] Theodora Anastasiou, Ioannis Pastellas, Sophia Karagiorgou, "Adversarial Explanations for Informed Civilian and Environmental Protection," 2024 IEEE International Conference on Big Data (BigData), Washington, DC, USA, 2024, pp. 2672-2681, doi: 10.1109/BigData62323.2024.10825734. link
[2] Theodora Anastasiou, Ioannis Pastellas, Sophia Karagiorgou, Mariza Konidi, “Explanation- Driven Adversarial Attacks against Multimedia Edge Applications”, 2025 6th International Conference in Electronic Engineering & Information Technology (EEITE), Chania, Greece, 2025. link
[3] Ioannis Pastellas, Sophia Karagiorgou, Mariza Konidi, “Adaptive Policy-Driven Network Intelligence for Edge-to-Cloud Continuum,” 31st ICE IEEE/ITMC Conference (ICE 2025), Valencia – Spain. link
[4] Under Review: P. Papaioannou, I. Pastellas, S. Karagiorgou, C. Tranoris, S. Denazis. AI-fuelled Adaptive Handovers and Energy-aware Mechanisms in 5G Cellular Networks. 2025 IEEE Conference on Network Function Virtualization and Software Defined Networks (NFV-SDN). link
