Spoofing features embedded into live face images with a de-identification model create visually clean poisoned training data that makes a face anti-spoofing model accept a specific spoofing attack as live.
Learning deep models fo r face anti-spoofing: Binary or auxiliary supervision,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Backdoor Poisoning Attack Against Face Spoofing Attack Detection Methods
Spoofing features embedded into live face images with a de-identification model create visually clean poisoned training data that makes a face anti-spoofing model accept a specific spoofing attack as live.