Skip to main content
. 2018 Jul 17;18(7):2312. doi: 10.3390/s18072312

Figure 10.

Figure 10

(a) Sequences from SynVideo2; and (b) SSIM with ground-truth for each magnified frame (top). Here, SSIM is calculated only by the pixels inside the red rectangle (see (a)). Similarly, PSNR and rMSE are measured by the same sites included (a), as shown in (c,d). SSIM shows that our algorithm has better structural similarity. PSNR and rMSE imply that our algorithm can magnify small signals with few magnification artifacts over other examined techniques.