Image Matting On Am 2K
평가 지표
MAD
MSE
SAD
평가 결과
이 벤치마크에서 각 모델의 성능 결과
모델 이름 | MAD | MSE | SAD | Paper Title | Repository |
|---|---|---|---|---|---|
| LF | 0.0210 | 0.0116 | 36.12 | A Late Fusion CNN for Digital Matting | |
| GFM(r) | 0.0064 | 0.0029 | 10.89 | Bridging Composite and Real: Towards End-to-end Deep Image Matting | |
| GFM(r2b) | 0.0060 | 0.0028 | 10.24 | Bridging Composite and Real: Towards End-to-end Deep Image Matting | |
| GFM(d) | 0.0059 | 0.0029 | 10.26 | Bridging Composite and Real: Towards End-to-end Deep Image Matting | |
| SHM | 0.0102 | 0.0068 | 17.81 | Semantic Human Matting | |
| GFM(r') | 0.0056 | 0.0024 | 9.66 | Bridging Composite and Real: Towards End-to-end Deep Image Matting | |
| SHMC | 0.0356 | 0.0270 | 61.50 | Boosting Semantic Human Matting with Coarse Annotations | |
| StyleMatte | 0.0055 | 0.0024 | 9.602 | Adversarially-Guided Portrait Matting | |
| HATT | 0.0161 | 0.0055 | 28.01 | Attention-Guided Hierarchical Structure Aggregation for Image Matting |
0 of 9 row(s) selected.