Nvidia announced updates to its AI media tools at the IBC conference in Amsterdam, running from September 11 to 14. Nvidia reported that the expansion covers software development kits, microservices, playbooks, and blueprints designed to enhance audio, video, and augmented-reality effects.
The company highlighted the Synthetic Video Detector microservice, which assesses whether footage is authentic or generated by AI. Nvidia stated that the detector reaches an accuracy of 99.3 percent for text-to-video content and 97.7 percent for image-to-video content.
Platform Integrations
Dalet is integrating the detector into a secure cloud workflow for news organizations. TwelveLabs added frame-level authenticity signals to its compliance platform. Wowza plans to distribute the detector through its video infrastructure framework across more than 35,000 video deployments.
Nvidia also introduced the 3D Body Pose tool, which estimates joint locations from a single camera. Vizrt uses this technology in live virtual studios to drive real-time 3D lighting effects. Ross Video is integrating Video Frame Generation into its Rio Replay platform to support slow-motion video creation for sports production.
