Scoop: Mitigation of Recapture Attacks on Provenance-Based Media Authentication
Yuxin (Myles) Liu
34th USENIX Security Symposium (USENIX Security '25) · Day 2 · System Security 3: Mobile Platforms
In an era dominated by rapidly spreading digital information and the proliferation of sophisticated generative AI, distinguishing authentic content from fabricated material has become an increasingly critical challenge. This talk, "Scoop: Mitigation of Recapture Attacks on Provenance-Based Media Authentication," by Yuxin (Myles) Liu and collaborators from UC Irvine and Microsoft, addresses a significant vulnerability in emerging **provenance-based media authentication** systems. While these systems cryptographically prove the origin and history of digital content, they are inherently blind to manipulations that occur *within* the content itself, particularly through **recapture attacks**.
AI review
Scoop attacks a real and underappreciated gap in provenance systems — the fact that C2PA/CAI authenticate the chain of custody, not the content itself — and proposes a physically grounded countermeasure that doesn't play the eternal cat-and-mouse game of artifact detection. The depth-discrepancy approach is elegant: ToF gives you ground-truth geometry; monocular depth estimation gives you what a 3D scene should look like; disagreement exposes the flat screen masquerading as reality. That's a clean, novel framing.