Abstract
Monocular face reconstruction is a significant task in many multimedia applications. However, learning-based methods unequivocally suffer from the lack of large datasets annotated with 3D ground-truth. To tackle this problem, we proposed a novel end-to-end 3D face reconstruction network consisting of a domain-transfer conditional GAN (cGAN) and a face reconstruction network. Our method first uses cGAN to translate the realistic face images to the specific rendered style, with a novel 2D facial edge consistency loss function to exploit in-the-wild images. The domain-transferred images are then fed into 3D face reconstruction network. We further propose a novel reprojection consistency loss to restrict 3D face reconstruction network in a self-supervised way. Our approach can be trained with annotated dataset, synthetic dataset and in-the-wild images to learn a unified face model. Extensive experiments have demonstrated the effectiveness of our method.
Funding Information
  • National Key Research and Development Plan of China (2020YFC2006000)
  • National Natural Science Foundation of China (72071089)