GenStream: Semantic Streaming Framework for Generative Reconstruction of Human-centric Media
ACM Multimedia 2025
October 27 – October 31, 2025
Dublin, Ireland
[PDF]
Emanuele Artioli (AAU, Austria), Daniele Lorenzi (AAU, Austria), Shivi Vats (AAU, Austria),Farzad Tashtarian (AAU, Austria), Christian Timmerer (AAU, Austria)
Abstract: Video streaming dominates global internet traffic, yet conventional pipelines remain inefficient for structured, human-centric content such as sports, performance, or interactive media. Standard codecs re-encode entire frames, foreground and background alike, treating all pixels uniformly and ignoring the semantic structure of the scene. This leads to significant bandwidth waste, particularly in scenarios where backgrounds are static and motion is constrained to a few salient actors. We introduce GenStream, a semantic streaming framework that replaces dense video frames with compact, structured metadata. Instead of transmitting pixels, GenStream encodes each scene as a combination of skeletal keypoints, camera viewpoint parameters, and a static 3D background model. These elements are transmitted to the client, where a generative model reconstructs photorealistic human figures and composites them into the 3D scene from the original viewpoint. This paradigm enables extreme compression, achieving over 99.9% bandwidth reduction compared to HEVC. We partially validate GenStream on Olympic figure skating footage and demonstrate potential high perceptual fidelity under minimal data. Looking forward, GenStream opens new directions in volumetric avatar synthesis, canonical 3D actor fusion across views, personalized and immersive viewing experiences at arbitrary viewpoints, and lightweight scene reconstruction, laying the groundwork for scalable, intelligent streaming in the post-codec era.
