Image Matting On Composition 1K 1
评估指标
Conn
Grad
MSE
SAD
评测结果
各个模型在此基准测试上的表现结果
| Paper Title | Repository | |||||
|---|---|---|---|---|---|---|
| DIM | 50.8 | 31.0 | 14.0 | - | Deep Image Matting | |
| PP-Matting | 45.4 | 22.69 | 5.0 | 46.22 | PP-Matting: High-Accuracy Natural Image Matting | |
| IndexNet-Matting | 43.7 | 25.9 | 13.0 | 45.8 | Indices Matter: Learning to Index for Deep Image Matting | |
| Context-Aware Matting | 33.2 | 17.3 | 8.2 | 35.8 | Context-Aware Image Matting for Simultaneous Foreground and Alpha Estimation | |
| SIM | 24.8 | 10.8 | 5.8 | 28.0 | Semantic Image Matting | |
| FBAMatting | 21.5 | 10.6 | 5.3 | 26.4 | $F$, $B$, Alpha Matting | |
| LSAMatting | 21.5 | 9.25 | 5.4 | 25.9 | Improving Deep Image Matting via Local Smoothness Assumption | |
| MatteFormer | 18.9 | 8.7 | 4.0 | 23.8 | MatteFormer: Transformer-Based Image Matting via Prior-Tokens | |
| LFPNet | 18.5 | 8.4 | 4.1 | 23.6 | Long-Range Feature Propagating for Natural Image Matting | |
| TMFNet | 17.6 | 6.7 | 3.6 | 22.1 | Trimap-guided Feature Mining and Fusion Network for Natural Image Matting | |
| ViTMatte | 14.78 | 6.74 | 3.0 | 20.33 | ViTMatte: Boosting Image Matting with Pretrained Plain Vision Transformers | |
| AEMatter | 12.46 | 4.76 | 2.26 | 17.53 | Revisiting Context Aggregation for Image Matting | |
| DiffMatte | 11.42 | 5.13 | 2.26 | 17.15 | Diffusion for Natural Image Matting |
0 of 13 row(s) selected.