Image Super Resolution On Manga109 4X
评估指标
PSNR
SSIM
评测结果
各个模型在此基准测试上的表现结果
比较表格
| 模型名称 | PSNR | SSIM |
|---|---|---|
| memnet-a-persistent-memory-network-for-image | 29.42 | 0.8942 |
| ml-craist-multi-scale-low-high-frequency | 31.17 | 0.9176 |
| esrgan-enhanced-super-resolution-generative | 24.89 | 0.7866 |
| esrgan-enhanced-super-resolution-generative | 31.66 | 0.9196 |
| fast-accurate-and-lightweight-super-1 | 30.40 | 0.9082 |
| deep-back-projection-networks-for-single | 31.74 | 0.921 |
| mair-a-locality-and-continuity-preserving | 32.66 | 0.9297 |
| swinfir-revisiting-the-swinir-with-fast | 33.03 | - |
| swinir-image-restoration-using-swin | 32.22 | 0.9273 |
| dual-aggregation-transformer-for-image-super | 32.67 | 0.9301 |
| second-order-attention-network-for-single | 31.66 | 0.9222 |
| learning-from-history-task-agnostic-model | 31.75 | 0.9229 |
| residual-dense-network-for-image-super | 31.0 | 0.9151 |
| image-super-resolution-via-attention-based | 31.79 | 0.921 |
| recursive-generalization-transformer-for | 32.68 | 0.9303 |
| hmanet-hybrid-multi-axis-aggregation-network | 33.19 | 0.9344 |
| gated-multiple-feedback-network-for-image | 31.24 | 0.9174 |
| image-super-resolution-using-deep | 27.58 | 0.8555 |
| image-super-resolution-using-very-deep | 31.22 | 0.9173 |
| densely-residual-laplacian-super-resolution | 31.78 | 0.9211 |
| perception-oriented-single-image-super | 28.08 | 0.8554 |
| accurate-image-super-resolution-using-very | 28.83 | 0.8870 |
| channel-partitioned-windowed-attention-and | 32.85 | 0.9318 |
| feedback-network-for-image-super-resolution | 31.15 | 0.9160 |
| activating-more-pixels-in-image-super | 33.09 | 0.9335 |
| auto-encoded-supervision-for-perceptual-image | 30.061 | 0.888 |
| hierarchical-information-flow-for-generalized | 33.13 | 0.9366 |
| accelerating-the-super-resolution | 27.90 | 0.8610 |
| transforming-image-super-resolution-a | 30.72 | 0.9111 |
| progressive-perception-oriented-network-for | 31.59 | - |
| drct-saving-image-super-resolution-away-from | 33.14 | 0.9347 |
| drct-saving-image-super-resolution-away-from | 32.96 | 0.9324 |
| progressive-perception-oriented-network-for | - | 0.9211 |
| deep-back-projection-networks-for-super | - | 0.914 |
| efficient-long-range-attention-network-for | 31.68 | 0.9226 |
| activating-more-pixels-in-image-super | 32.87 | 0.9319 |
| residual-feature-aggregation-network-for | 31.41 | 0.9187 |
| mair-a-locality-and-continuity-preserving | 32.46 | 0.9284 |
| lightweight-image-super-resolution-with-1 | 30.45 | - |
| lightweight-feature-fusion-network-for-single | 29.76 | 0.8979 |
| dual-aggregation-transformer-for-image-super | 32.51 | 0.9291 |
| single-image-super-resolution-via-a-holistic | 31.73 | 0.9207 |
| swinfir-revisiting-the-swinir-with-fast | 32.83 | 0.9314 |
| progressive-multi-scale-residual-network-for | 31.07 | 0.9144 |
| ml-craist-multi-scale-low-high-frequency | 31.11 | 0.9162 |
| hierarchical-back-projection-network-for | 31.57 | 0.92 |
| enhanced-deep-residual-networks-for-single | 31.02 | 0.9148 |
| recursive-generalization-transformer-for | 32.50 | 0.9291 |
| image-super-resolution-with-cross-scale-non | 31.43 | 0.9201 |