Semantic Segmentation on Replica (novel views)
92mIoUFSTM
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| FSTM2026.05 | 92 | — | — | — | |
| RICO2026.05 | 86.2 | — | — | — | |
| ObjectSDF++2026.05 | 83.6 | — | — | — | |
| Feature-3DGS2026.06 | 78.9 | 94.3 | 629.7 | 3 | |
| MLP-Splatting2026.06 | 77.7 | 93.9 | 32 | 8 | |
| Nerf-DFF2026.06 | 63.6 | 86.4 | 86.5 | 3 | |
| EPS3DNumber of views=82026.06 | 48.33 | 85.12 | — | — | |
| LSegNumber of views=82026.06 | 40.38 | 82.24 | — | — | |
| Feature-3DGSNumber of views=82026.06 | 32.6 | 79.32 | — | — | |
| Uni3RNumber of views=82026.06 | 32.16 | 74.2 | — | — | |
| OursSemantic encoder=CLIP-LSeg, Training views=122025.03 | 29.173 | 75.482 | — | — | |
| LSMNumber of views=82026.06 | 28.74 | 67.87 | — | — | |
| FSGSSemantic encoder=CLIP-LSeg, Training views=122025.03 | 23.721 | 71.235 | — | — | |
| DNGaussianSemantic encoder=CLIP-LSeg, Training views=122025.03 | 23.692 | 70.705 | — | — | |
| CoR-GSSemantic encoder=CLIP-LSeg, Training views=122025.03 | 22.496 | 68.901 | — | — | |
| Gau-GroupingSemantic encoder=CLIP-LSeg, Training views=122025.03 | 20.252 | 66.519 | — | — | |
| Feature 3DGSSemantic encoder=CLIP-LSeg, Training views=122025.03 | 16.357 | 64.955 | — | — | |
| 3DOVSSemantic encoder=CLIP-LSeg, Training views=122025.03 | 8.859 | 41.909 | — | — |