首页|Patent Issued for Utilizing multiple stacked machine learning models to detect d eepfake content (USPTO 12062105)
Patent Issued for Utilizing multiple stacked machine learning models to detect d eepfake content (USPTO 12062105)
扫码查看
点击上方二维码区域,可以放大扫码查看
原文链接
NETL
NSTL
Reporters obtained the following quote from the background information supplied by the inventors:“Digital content (e.g., images, audio, video, and/or the like) can be manipulated with artificial intelligencebasedsolutions. For example, a n image (e.g., a company’s design sketches) can be manipulated byartificial int elligence (e.g., a deepfake model) to generate substantially similar images (e.g ., designs)that infringe the company’s design sketches. Audio can be manipulate d by artificial intelligence for voiceimpersonation. Moreover, images and video s can be combined and superimposed onto source images orvideos using artificial intelligence.”
Accenture Global Solutions LimitedArti ficial IntelligenceBusinessCyborgsEmerging TechnologiesMachine Learning