Video Super Resolution On Msu Video Upscalers
المقاييس
LPIPS
PSNR
SSIM
النتائج
نتائج أداء النماذج المختلفة على هذا المعيار القياسي
جدول المقارنة
| اسم النموذج | LPIPS | PSNR | SSIM |
|---|---|---|---|
| swinir-image-restoration-using-swin | 0.183 | 28.86 | 0.830 |
| esrgan-enhanced-super-resolution-generative | - | 27.29 | 0.936 |
| local-global-fusion-network-for-video-super | - | 27.42 | 0.939 |
| النموذج 4 | 0.292 | 31.01 | 0.859 |
| real-time-single-image-and-video-super | - | 26.25 | 0.926 |
| النموذج 6 | 0.206 | 31.00 | 0.890 |
| النموذج 7 | 0.218 | 30.55 | 0.868 |
| النموذج 8 | 0.210 | 30.55 | 0.869 |
| real-esrgan-training-real-world-blind-super | 0.185 | 28.82 | 0.850 |
| النموذج 10 | 0.369 | 30.10 | 0.854 |
| النموذج 11 | 0.212 | 29.01 | 0.822 |
| deep-blind-video-super-resolution | - | 27.28 | 0.937 |
| النموذج 13 | 0.195 | 31.15 | 0.898 |
| basicvsr-improving-video-super-resolution | 0.334 | 30.98 | 0.881 |
| iseebetter-spatio-temporal-video-super | - | 27.42 | 0.939 |
| النموذج 16 | 0.278 | 30.76 | 0.838 |
| designing-a-practical-degradation-model-for | 0.177 | 29.27 | 0.836 |
| learning-a-single-convolutional-super | 0.349 | 30.96 | 0.852 |
| swinir-image-restoration-using-swin | 0.189 | 28.55 | 0.845 |
| real-esrgan-training-real-world-blind-super | 0.181 | 29.14 | 0.855 |
| real-time-super-resolution-system-of-4k-video | - | 26.33 | 0.929 |
| real-time-video-super-resolution-with-spatio | - | 26.92 | 0.932 |
| dynavsr-dynamic-adaptive-blind-video-super | - | 26.12 | 0.916 |
| النموذج 24 | 0.219 | 30.65 | 0.836 |
| comisr-compression-informed-video-super | 0.291 | 30.97 | 0.871 |
| real-esrgan-training-real-world-blind-super | 0.244 | 28.71 | 0.830 |
| vrt-a-video-restoration-transformer | 0.343 | 31.01 | 0.869 |
| النموذج 28 | 0.316 | 30.13 | 0.886 |
| النموذج 29 | 0.249 | 29.48 | 0.825 |
| النموذج 30 | 0.178 | 31.28 | 0.879 |
| النموذج 31 | 0.423 | 30.75 | 0.833 |
| frame-recurrent-video-super-resolution | - | 27.23 | 0.936 |
| النموذج 33 | 0.349 | 27.44 | 0.838 |
| designing-a-practical-degradation-model-for | 0.301 | 30.19 | 0.859 |
| accurate-image-super-resolution-using-very | - | 25.89 | 0.917 |
| image-super-resolution-via-deep-recursive | - | 26.97 | 0.933 |
| temporally-coherent-gans-for-video-super | - | 26.60 | 0.933 |
| real-world-super-resolution-via-kernel | 0.220 | 30.64 | 0.900 |
| deep-video-super-resolution-using-hr-optical | - | 27.14 | 0.937 |
| real-esrgan-training-real-world-blind-super | 0.333 | 25.52 | 0.795 |
| real-esrgan-training-real-world-blind-super | 0.296 | 30.52 | 0.878 |
| real-esrgan-training-real-world-blind-super | 0.280 | 30.01 | 0.868 |
| النموذج 43 | 0.188 | 31.14 | 0.888 |
| detail-revealing-deep-video-super-resolution | - | 26.99 | 0.933 |
| investigating-tradeoffs-in-real-world-video | 0.201 | 29.54 | 0.838 |
| النموذج 46 | 0.244 | 30.50 | 0.851 |
| النموذج 47 | 0.313 | 31.12 | 0.853 |
| image-super-resolution-using-deep | - | 26.68 | 0.929 |