Audio-visual occlusion-robust gender recognition and age estimation approach based on multi-task cross-modal attention

Maxim Markitantov, Elena Ryumina, Alexey Karpov

Abstract: Highlights•A novel unimodal and bimodal transformer-based approach for multi-task gender, age and mask type prediction.•A study of protective mask type influence on gender recognition and age estimation.•A cross-corpus study to improve the generalizability of the proposed approach.•Application of a transformer-based approach for biometric characteristics’ recognition in HR expert systems.