AI4Media@AI4EU Café: Video Compression turns to Artificial Intelligence
Speaker: Roberto Iacoviello (Lead Research Engineer at RAI R&D)
Artificial Intelligence (AI) technologies are used in more and more applications yielding one of the fastest growing research in data analysis. Their usage can now be found in wide range of applications, including Image Processing, Image Enhancement and, recently, Video Coding. Users demand high resolutions, wide range of colors and richer details under bandwidth constraint. For this purpose, current state of the art image and video compression algorithms (e.g. High Efficiency Video Coding - HEVC - and its successor Versatile Video Coding - VVC) have some limitations: they are agnostic to the semantics of data being compressed as they work at low level (i.e., pixel level), and they perform exhaustive search for rate-distortion optimization. In order to produce more accurate and visually pleasing pictures at much higher compression levels, new generations of AI-empowered compression standards are needed.
Roberto Iacoviello graduated from the Politecnico di Torino. Since 2007 he has been working at the Rai R&D where he has developed the following skills: End to end video compression for broadcast, Computer Vision Algorithms, Augmented and Virtual Reality, Multimodal interfaces for TV, Smartphones and Head Mounted Display. His current research focuses on Artificial Intelligence applied to video compression and point clouds, next-generation TV Services, in particular Augmented Reality in broadcast television. He follows the activities of the European Broadcasting Union (EBU), Digital Video Broadcasting (DVB) and actively participates in MPEG (Moving Picture Experts Group) and MPAI (Moving Picture, Audio and Data Coding by Artificial Intelligence) meetings. He has been working on several EC funded projects among which 5G-City and, at present, 5G-TOURS and AI4Media.