Table 2.
Average evaluation metric values of different methods on LLVIP dataset. The best value in each metric is denoted in bold, and the second-best score is highlighted with an underline.
| Methods | EN | SF | SD | AG | Deep Learning | |
|---|---|---|---|---|---|---|
| Wavelet [29] | 6.8964 | 0.0241 | 9.4977 | 0.1996 | 2.0167 | ✗ |
| FPDE [33] | 6.9161 | 0.0451 | 9.4264 | 0.4909 | 3.6297 | ✗ |
| ADF [30] | 6.9282 | 0.0489 | 9.4236 | 0.5273 | 3.8861 | ✗ |
| LatLRR [32] | 6.9748 | 0.0450 | 9.3101 | 0.4535 | 3.2089 | ✗ |
| TIF [31] | 7.0605 | 0.0635 | 9.4683 | 0.6354 | 4.7440 | ✗ |
| IFEVIP [34] | 7.4487 | 0.0566 | 9.6836 | 0.4957 | 4.1067 | ✗ |
| DenseFuse [8] | 6.8899 | 0.0375 | 9.4237 | 0.3530 | 2.9379 | ✓ |
| FusionGAN [6] | 7.0468 | 0.0293 | 10.0528 | 0.2956 | 2.3374 | ✓ |
| TarDal [5] | 7.1872 | 0.0511 | 9.6212 | 0.3857 | 3.5221 | ✓ |
| STDFusionNet [2] | 5.4825 | 0.0522 | 6.8897 | 0.4898 | 3.4384 | ✓ |
| DIDFuse [10] | 6.1477 | 0.0508 | 8.0359 | 0.3605 | 6.2487 | ✓ |
| SeAFusion [7] | 7.4457 | 0.0626 | 9.8828 | 0.6254 | 4.7663 | ✓ |
| DIVFusion [28] | 7.5716 | 0.0547 | 10.0577 | 0.3312 | 4.6006 | ✓ |
| Ours | 7.6913 | 0.0667 | 10.1742 | 0.4865 | 5.6376 | ✓ |