SER-FIQ: Unsupervised Estimation of Face Image Quality Based on Stochastic Embedding Robustness

Visualization of the proposed unsupervised face quality assessment concept. We propose using the robustness of an image representation as a quality clue. Our approach defines this robustness based on the embedding variations of random subnetworks of a given face recognition model. An image that produces small variations in the stochastic embeddings (bottom left), demonstrates high robustness (red areas on the right) and thus, high image quality. Contrary, an image that produces high variations in the stochastic embeddings (top left) coming from random subnetworks, indicates a low robustness (blue areas on the right). Therefore, it is considered as low quality.