Semantic Segmentation On Stanford2D3D Rgbd
評価指標
Pixel Accuracy
mIoU
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 | Pixel Accuracy | mIoU | Paper Title | Repository |
|---|---|---|---|---|
| CMX (SegFormer-B2) | 82.3 | 61.2 | CMX: Cross-Modal Fusion for RGB-X Semantic Segmentation with Transformers | |
| ShapeConv-101 | 82.7 | 60.6 | ShapeConv: Shape-aware Convolutional Layer for Indoor RGB-D Semantic Segmentation | |
| CMX (SegFormer-B4) | 82.6 | 62.1 | CMX: Cross-Modal Fusion for RGB-X Semantic Segmentation with Transformers | |
| Depth-aware CNN | 65.4 | 39.5 | Depth-aware CNN for RGB-D Segmentation | |
| MMAF-Net-152 | 76.5 | 52.9 | Multi-Modal Attention-based Fusion Model for Semantic Segmentation of RGB-Depth Images | - |
| Linear Fusion (Segformer B2) | - | 57.16 | Missing Modality Robustness in Semi-Supervised Multi-Modal Semantic Segmentation |
0 of 6 row(s) selected.