Video Inpainting On Davis
평가 지표
LPIPS (object)
LPIPS (square)
PNSR (object)
SSIM (object)
SSIM (square)
평가 결과
이 벤치마크에서 각 모델의 성능 결과
모델 이름 | LPIPS (object) | LPIPS (square) | PNSR (object) | SSIM (object) | SSIM (square) | Paper Title | Repository |
|---|---|---|---|---|---|---|---|
| FGT++ | 0.035 | 0.028 | 35.61 | 0.961 | 0.971 | Exploiting Optical Flow Guidance for Transformer-Based Video Inpainting | |
| VINet | - | - | - | - | - | Deep Video Inpainting | |
| FGVC | - | - | - | - | - | Flow-edge Guided Video Completion | |
| DFVI | - | - | - | - | - | Deep Flow-Guided Video Inpainting | |
| FuseFormer | - | - | - | - | - | FuseFormer: Fusing Fine-Grained Information in Transformers for Video Inpainting | |
| E2FGVI | - | - | - | - | - | Towards An End-to-End Framework for Flow-Guided Video Inpainting | |
| DMT | - | - | - | - | - | Deficiency-Aware Masked Transformer for Video Inpainting | |
| CAP | - | - | - | - | - | Copy-and-Paste Networks for Deep Video Inpainting | |
| STTN | - | - | - | - | - | Learning Joint Spatial-Temporal Transformations for Video Inpainting | |
| LGTSM | - | - | - | - | - | Learnable Gated Temporal Shift Module for Deep Video Inpainting | |
| FGT++* | 0.027 | 0.022 | 35.9 | 96.8 | 97.6 | Exploiting Optical Flow Guidance for Transformer-Based Video Inpainting |
0 of 11 row(s) selected.