Ensuring privacy in face recognition: a survey on data generation, inference and storage
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
As facial recognition technology plays an increasingly pivotal role in biometric authentication, its potential threats to individual privacy have raised significant societal concerns. This paper provides a survey of privacy-preserving techniques across the three critical stages of facial recognition: data generation, model inference, and data storage. We explore challenges and methodologies for safeguarding privacy within facial recognition systems, given growing concerns over biometric data misuse. In particular, we highlight the shift from traditional datasets to synthetic counterparts, leveraging generative models like GANs and diffusion models to create diverse and realistic facial imagery without compromising privacy. At the model inference stage, we discuss privacy-preserving approaches, including transformation-based methods and cryptographic techniques such as homomorphic encryption. Finally, we examine the vulnerabilities of face templates and the cryptographic protections against inversion attacks. Our survey underscores the importance of balancing recognition accuracy with privacy preservation and calls for concerted research and policy efforts to advance privacy-centric face recognition technologies that respect individual rights while maintaining operational efficacy.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it