VIGFace: Virtual Identity Generation Model for Face Image Synthesis

arXiv:2403.08277v1 Announce Type: new
Abstract: Deep learning-based face recognition continues to face challenges due to its reliance on huge datasets obtained from web crawling, which can be costly to gather and raise significant real-world privacy concerns. To address this issue, we propose VIGFace, a novel framework capable of generating synthetic facial images. Initially, we train the face recognition model using a real face dataset and create a feature space for both real and virtual IDs where virtual prototypes are orthogonal …

abstract arxiv challenges concerns crawling cs.cv datasets deep learning face face recognition framework gather identity image issue novel privacy raise recognition reliance synthesis synthetic type virtual web web crawling world world privacy