Reproduce “Ghost Attack: Using 19th-Century Illusions to Fool 21st-Century Biometrics” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Tyler Fitzsimmons, Adam Czajka
Ghost Attack blends a displayed target face with a face image using reflected light before capture. The authors evaluate this optical mixture against open-source face matchers, presentation-attack detection, and morph-attack detection in a controlled lab. At a 5% false-match-rate threshold, ArcFace accepts 32.75% of attack frames; results vary by threshold and do not establish performance for commercial systems or human subjects.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.