Visual DNA is an exciting new AI technology, unique to Mobius Labs, that is already transforming the way we think about video ingest, tagging, search and archive.

At NAB 2023 we are proud to introduce a suite of new complementary audio tools that can generate transcripts with state-of-the-art capabilities and keywords based on spoken content of the video that will further supercharge your asset management system and the value of your content.

Mobius’ unique one-pass process creates an ultra-compact description of video frames (in essence, the Visual DNA) that is around 1,500 times smaller than the corresponding 4k video content. In parallel it can also generate highly accurate audio transcripts in 30+ languages, translate them into English and detect the keywords from the spoken content of the video without having to rely on separate natural language processing modules as required by traditional cloud-based vendors.

With Visual DNA, one hour of video content requires just 30MB of metadata storage, and can be re-indexed in less than 10 seconds. Thus, video that has been poorly tagged or has no metadata at all becomes a thing of the past and your content library can always be kept fresh and relevant.

Why not visit us at NAB on booth W1578 to see how we can literally turn back time on badly indexed archives and discover the many other benefits of our visual and audio solutions including sophisticated, granular AI tagging and face recognition with advanced face feature analysis.