Semantic Segmentation On Stanford2D3D 1
평가 지표
mAcc
mIoU
평가 결과
이 벤치마크에서 각 모델의 성능 결과
비교 표
| 모델 이름 | mAcc | mIoU |
|---|---|---|
| panoformer-panorama-transformer-for-indoor | 64.5 | 48.9% |
| sgat4pass-spherical-geometry-aware | - | 56.4% |
| interpolated-selectionconv-for-spherical | - | 41.4% |
| bending-reality-distortion-aware-transformers | - | 52.1% |
| single-frame-semantic-segmentation-using | 63.96 | 52.87% |
| spherical-cnns-on-unstructured-grids | 54.65 | 38.3% |
| sgat4pass-spherical-geometry-aware | - | 55.3% |
| tangent-images-for-mitigating-spherical | 65.2 | 45.6% |
| spin-weighted-spherical-cnns | - | 43.4% |
| gauge-equivariant-convolutional-networks-and | 55.9 | 39.4% |
| single-frame-semantic-segmentation-using | 68.79 | 58.24% |
| bending-reality-distortion-aware-transformers | - | 53.0% |
| distortion-aware-convolutional-filters-for | - | 34.6% |
| single-frame-semantic-segmentation-using | 70.68 | 60.6% |
| single-frame-semantic-segmentation-using | 69.03 | 59.43% |
| complementary-bi-directional-feature | 65.6 | 52.2% |
| hohonet-360-indoor-holistic-understanding | 65.0 | 52.0% |
| behind-every-domain-there-is-a-shift-adapting | - | 52.3% |
| single-frame-semantic-segmentation-using | 68.57 | 55.49% |
| bending-reality-distortion-aware-transformers | - | 51.2% |
| bending-reality-distortion-aware-transformers | - | 50.8% |
| orientation-aware-semantic-segmentation-on | 58.6 | 43.3% |
| bending-reality-distortion-aware-transformers | - | 48.1% |
| fredsnet-joint-monocular-depth-and-semantic | 63.1 | 46.1% |
| behind-every-domain-there-is-a-shift-adapting | - | 54.0% |