Text Image Super-Resolution by Image Matting and Text Label Supervision

These days, many methods have been proposed to deal with nature image super-resolution (SR) and get impressive performance. However, these methods don't do well in text images SR due to their ignorance of the difference between nature images and text images. In this paper, we propose a matting-based dual generative adversarial network (mdGAN) for text image SR. Firstly, the input image is decomposed into text, foreground and background layers using deep image matting. Then two parallel branches are constructed to recover text boundary information and color information respectively. Furthermore, in order to improve the restoration accuracy of characters in output image, we use the input image's corresponding ground truth text label as extra supervise information to refine the two-branch networks during training. Experiments on real text images demonstrate that our method outperforms several state-of-the-art methods quantitatively and qualitatively.