Wireva

UCLA Researchers Use Light-Based Computing to Detect Deepfake Videos

A hybrid optical-neural system developed at UCLA can identify AI-generated deepfake videos with up to 97.8% accuracy while processing multiple video streams simultaneously and consuming less energy than conventional digital detectors.

To detect deepfake videos, researchers turn to the physics of light
UCLA Researchers Use Light-Based Computing to Detect Deepfake Videos
Deepfake · Wikimedia — licence per file · rights
Monitoring item. The full text is not distributed. Extract and source below.

A hybrid optical-neural system developed at UCLA can identify AI-generated deepfake videos with up to 97.8% accuracy while processing multiple video streams simultaneously and consuming less energy than conventional digital detectors.

Same event, other desks

Story file →