How Private is Private? A Comparative Study for Face De-Identification
On the metric side, we propose HiFD, a Hierarchical Face De-identification metric that unifies identity suppression, multi-level utility preservation, and image quality under a single consistency-based paradigm: every component is computed from pretrained estimators' outputs on the original face and its de-identified counterpart, directly quantifying how much identity is suppressed and how much downstream-perceivable utility survives.
ProofPaper ↗
Key points
- Face de-identification (FDeID) has emerged as a critical privacy-preserving technology, yet its evaluation remains fundamentally fragmented.
- We revisit FDeID evaluation from both the data and metric perspectives.
- On the data side, we introduce UtilFace, a curated, demographically balanced benchmark with high identity diversity, assembled from four large-scale face datasets through identity-aware cleaning, resolution enhancement, and stratified filtering.
- We release the benchmark and evaluation toolkit to foster systematic and reproducible research in privacy-preserving human face analysis.
Sources (1)
- [1]How Private is Private? A Comparative Study for Face De-IdentificationarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 7, 04:16 PM
On the metric side, we propose HiFD, a Hierarchical Face De-identification metric that unifies identity suppression, multi-level utility preservation, and image quality under a single consistency-based paradigm: every component is computed from pretrained estimators' outputs on the original face and its de-identified counterpart, directly quantifying how much identity is suppressed and how much do
Face de-identification (FDeID) has emerged as a critical privacy-preserving technology, yet its evaluation remains fundamentally fragmented.
Extractive summary: sentences quoted from the sources.