Using five normalized facial landmarks and a Random Forest classifier, the paper reports 96.67% accuracy for assessing CCTV face quality and claims filtering low-quality faces cuts ArcFace false rejection rate from 13.19% to 0.04%.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
A Lightweight Face Quality Assessment Framework to Improve Face Verification Performance in Real-Time Screening Applications
Using five normalized facial landmarks and a Random Forest classifier, the paper reports 96.67% accuracy for assessing CCTV face quality and claims filtering low-quality faces cuts ArcFace false rejection rate from 13.19% to 0.04%.