Referring Image Matting Keyword Based On
評価指標
MAD
MAD(E)
MSE
MSE(E)
SAD
SAD(E)
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 | MAD | MAD(E) | MSE | MSE(E) | SAD | SAD(E) | Paper Title | Repository |
|---|---|---|---|---|---|---|---|---|
| CLIPSeg (ViT-B/16) | 0.0101 | 0.0106 | 0.0064 | 0.0067 | 17.75 | 18.69 | Image Segmentation Using Text and Image Prompts | |
| MDETR (ResNet-101) | 0.0183 | 0.0190 | 0.0137 | 0.0141 | 32.27 | 33.52 | MDETR -- Modulated Detection for End-to-End Multi-Modal Understanding | |
| CLIPMat (ViT-B/16) | 0.0057 | 0.0059 | 0.0028 | 0.0029 | 9.91 | 10.41 | Referring Image Matting | |
| CLIPMat (ViT-L/14) | 0.0049 | 0.0051 | 0.0022 | 0.0023 | 8.51 | 8.98 | Referring Image Matting |
0 of 4 row(s) selected.