Knowledge Priors for Identity-Disentangled Open-Set Privacy-Preserving Video FER
This addresses privacy concerns in video-based facial expression recognition for applications like surveillance or healthcare, but it is incremental as it builds on existing privacy-preserving methods.
The paper tackles the problem of privacy-preserving facial expression recognition in open-set video settings where identity labels are unavailable, by proposing a two-stage framework that uses knowledge priors to suppress identity and restore expression cues, achieving FER accuracy comparable to identity-supervised baselines on three datasets.
Facial expression recognition relies on facial data that inherently expose identity and thus raise significant privacy concerns. Current privacy-preserving methods typically fail in realistic open-set video settings where identities are unknown, and identity labels are unavailable. We propose a two-stage framework for video-based privacy-preserving FER in challenging open-set settings that requires no identity labels at any stage. To decouple privacy and utility, we first train an identity-suppression network using intra- and inter-video knowledge priors derived from real-world videos without identity labels. This network anonymizes identity while preserving expressive cues. A subsequent denoising module restores expression-related information and helps recover FER performance. Furthermore, we introduce a falsification-based validation method that uses recognition priors to rigorously evaluate privacy robustness without requiring annotated identity labels. Experiments on three video datasets demonstrate that our method effectively protects privacy while maintaining FER accuracy comparable to identity-supervised baselines.