Second FRCSyn-onGoing: Winning solutions and post-challenge analysis to improve face recognition with synthetic data

Ivan DeAndres-Tame, Ruben Tolosana, Pietro Melzi, Ruben Vera-Rodriguez, Minchul Kim, Christian Rathgeb, Xiaoming Liu, Luis F. Gomez, Aythami Morales, Julian Fierrez, Javier Ortega-Garcia, Zhizhou Zhong, Yuge Huang, Yuxi Mi, Shouhong Ding, Shuigeng Zhou, Shuai He, Lingzhi Fu, Heng Cong, Rongyu Zhang et al. (39 additional authors not shown)

Abstract: Highlights•Challenges in face recognition: Privacy, demographic bias, and generalization.•Synthetic data enables large-scale, customizable training for face recognition.•Exploration of synthetic data in constrained and unconstrained scenarios.•Novel Generative AI methods show improved performance in diverse scenarios.•Fusion of real and synthetic data, among others, can mitigate several challenges.

External IDs:doi:10.1016/j.inffus.2025.103099