Probabilistic Face Embeddings
This work addresses uncertainty in face recognition for risk-controlled systems, representing an incremental improvement over existing embedding methods.
The paper tackles the problem of noisy face representations in unconstrained settings by proposing Probabilistic Face Embeddings (PFEs), which model each face as a Gaussian distribution to incorporate uncertainty, leading to improved face recognition performance over deterministic embeddings.
Embedding methods have achieved success in face recognition by comparing facial features in a latent semantic space. However, in a fully unconstrained face setting, the facial features learned by the embedding model could be ambiguous or may not even be present in the input face, leading to noisy representations. We propose Probabilistic Face Embeddings (PFEs), which represent each face image as a Gaussian distribution in the latent space. The mean of the distribution estimates the most likely feature values while the variance shows the uncertainty in the feature values. Probabilistic solutions can then be naturally derived for matching and fusing PFEs using the uncertainty information. Empirical evaluation on different baseline models, training datasets and benchmarks show that the proposed method can improve the face recognition performance of deterministic embeddings by converting them into PFEs. The uncertainties estimated by PFEs also serve as good indicators of the potential matching accuracy, which are important for a risk-controlled recognition system.