Video Inpainting On Youtube Vos
評価指標
Ewarp
PSNR
SSIM
VFID
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
モデル名 | Ewarp | PSNR | SSIM | VFID | Paper Title | Repository |
|---|---|---|---|---|---|---|
| LGTSM | 0.1859 | 29.74 | 0.9504 | 0.070 | Learnable Gated Temporal Shift Module for Deep Video Inpainting | - |
| CAP | 0.1470 | 31.58 | 0.9607 | 0.071 | Copy-and-Paste Networks for Deep Video Inpainting | - |
| E2FGVI | 0.0864 | 33.71 | 0.9700 | 0.046 | Towards An End-to-End Framework for Flow-Guided Video Inpainting | - |
| FuseFormer | 0.0900 | 33.29 | 0.9681 | 0.053 | FuseFormer: Fusing Fine-Grained Information in Transformers for Video Inpainting | - |
| ProPainter | - | 34.43 | 0.9735 | 0.042 | ProPainter: Improving Propagation and Transformer for Video Inpainting | - |
| FGVC | 0.1022 | 29.67 | 0.9403 | 0.064 | Flow-edge Guided Video Completion | - |
| VINet | 0.1490 | 29.20 | 0.9434 | 0.072 | Deep Video Inpainting | - |
| DFVI | 0.1509 | 29.16 | 0.9429 | 0.066 | Deep Flow-Guided Video Inpainting | - |
| DMT | - | 34.27 | 0.9730 | 0.044 | Deficiency-Aware Masked Transformer for Video Inpainting | - |
| STTN | 0.0907 | 32.34 | 0.9655 | 0.053 | Learning Joint Spatial-Temporal Transformations for Video Inpainting | - |
0 of 10 row(s) selected.