In this session from IBC2024, three authors from NHK, Viaccess-Orca and European Broadcasting Union present their work on the application of AI to media production as part of the IBC Technical Papers.

Our first paper from the EBU describes a facial recognition machine learning system that has been tailored to the specific needs of media documentalists. It seeks to only identify “active” and not incidental characters, assisting in compliance with privacy regulations. Our second paper, from NHK, presents a solution to the over-training problem when applying machine learning to multi-label indexing of news items, this is especially important for infrequently classified events. It provides good performance and has been trialled, with the authors also acknowledging its limitations.  Our final paper, from Viaccess-Orca, looks to a future where Generative AI is used to adapt advertising based on characteristics / beliefs that have been gleaned of the target.  Used in a closed loop, the Generative AI can test, score and adapt the media to maximise its impact. 

Moderator

Stephan Heimbecher (SWR) 

Click below to access the following papers

Topics