AlphaGAN: Generative adversarial networks for natural image matting
This addresses the problem of accurate image matting for practical applications in film and TV production, representing an incremental advance over existing methods.
The authors tackled natural image matting by introducing the first GAN-based method, which predicts high-quality alpha mattes with improved handling of fine structures like hair, achieving state-of-the-art results on the alphamatting benchmark for gradient error.
We present the first generative adversarial network (GAN) for natural image matting. Our novel generator network is trained to predict visually appealing alphas with the addition of the adversarial loss from the discriminator that is trained to classify well-composited images. Further, we improve existing encoder-decoder architectures to better deal with the spatial localization issues inherited in convolutional neural networks (CNN) by using dilated convolutions to capture global context information without downscaling feature maps and losing spatial information. We present state-of-the-art results on the alphamatting online benchmark for the gradient error and give comparable results in others. Our method is particularly well suited for fine structures like hair, which is of great importance in practical matting applications, e.g. in film/TV production.