DisCO: Portrait Distortion Correction with Perspective-Aware 3D GANs

Zhixiang Wang, Yu-Lun Liu, Jia-Bin Huang, Shin'ichi Satoh, Sizhuo Ma, Guru Krishnan, Jian Wang
Event IJCV
Research Areas Computer Vision, Computational Photography

Close-up facial images captured at short distances often suffer from perspective distortion, resulting in exaggerated facial features and unnatural/unattractive appearances. We propose a simple yet effective method for correcting perspective distortions in a single closeup face image. We first perform 3D GAN inversion using a perspective-distorted input facial image by jointly optimizing the intrinsic and extrinsic camera parameters and the face latent code. To address the ambiguity inherent in this joint optimization, we develop starting from a short distance, optimization scheduling, reparametrizations, and geometric regularization. Re-rendering the portrait at a proper focal length and camera distance effectively corrects perspective distortions and produces more natural-looking results. We also incorporate a workflow to handle full images rather than limiting our method to cropped faces. Our experiments show that our method compares favorably against previous approaches qualitatively and quantitatively. We showcase numerous examples validating the applicability of our method on in-the-wild portrait photos. Our code is available at