This paper was converted on www.awesomepapers.org from LaTeX by an anonymous user.
Want to know more? Visit the Converter page.

11institutetext: Anonymous Organization

Supplemental

Anonymous

We present more qualitative results in the supplemental.

Refer to caption
(a) Test dataset: MURA-mini.
Refer to caption
(b) Test dataset: MURA-plus.
Figure 1: Qualitative comparisons on several representative samples with upsampling scale factor of 4 and 2. AID-SRGAN outperforms previous approaches in both removing artifacts and restoring texture details. The PSNR↑\uparrow/SSIM↑\uparrow of the test images are shown in the figure. The best results are in red. (Zoom in for best view.)
Refer to caption
Refer to caption
Figure 2: (a): Ablation studies on different modules and hyper-parameters in AID-SRGAN. The best SSEQ↓\downarrow (No Ref.) are in bold, and the second best results are underlined. (b): Examples of training samples and their corresponding LR images. (Zoom in for best view.)