Face Recognition
Face recognition research aims to develop accurate and robust systems for identifying individuals from their facial images. Current efforts focus on improving performance under challenging conditions (e.g., low-resolution images, occlusions), mitigating biases stemming from demographic imbalances in training data, and enhancing the explainability and security of these systems through techniques like knowledge distillation and adversarial watermarking. These advancements have significant implications for various applications, including security, law enforcement, and healthcare, while also raising important ethical considerations regarding privacy and fairness.
Papers
Assessing Bias in Face Image Quality Assessment
Žiga Babnik, Vitomir Štruc
Meet-in-the-middle: Multi-scale upsampling and matching for cross-resolution face recognition
Klemen Grm, Berk Kemal Özata, Vitomir Štruc, Hazım Kemal Ekenel
MixFairFace: Towards Ultimate Fairness via MixFair Adapter in Face Recognition
Fu-En Wang, Chien-Yi Wang, Min Sun, Shang-Hong Lai