SREFI: Synthesis of Realistic Example Face Images
This addresses the need for scalable and privacy-compliant face datasets for computer vision researchers, though it appears incremental as it builds on existing synthesis methods.
The paper tackles the problem of expanding face image datasets without identity-labeling and privacy issues by proposing a face synthesis approach that generates synthetic images of both real and synthetic identities. Experiments showed that training with augmented synthetic faces boosted face recognition performance, with a CNN model trained on nearly 200,000 synthetic faces improving recognition on challenging real images.
In this paper, we propose a novel face synthesis approach that can generate an arbitrarily large number of synthetic images of both real and synthetic identities. Thus a face image dataset can be expanded in terms of the number of identities represented and the number of images per identity using this approach, without the identity-labeling and privacy complications that come from downloading images from the web. To measure the visual fidelity and uniqueness of the synthetic face images and identities, we conducted face matching experiments with both human participants and a CNN pre-trained on a dataset of 2.6M real face images. To evaluate the stability of these synthetic faces, we trained a CNN model with an augmented dataset containing close to 200,000 synthetic faces. We used a snapshot of this trained CNN to recognize extremely challenging frontal (real) face images. Experiments showed training with the augmented faces boosted the face recognition performance of the CNN.