Scene Classification on NYU v2
80.3AccuracyOMNIVORE
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| OMNIVOREBackbone=Swin-L2022.01 | 80.3 | — | |
| OMNIVOREBackbone=Swin-B2022.01 | 80 | — | |
| Omnivore0-shot=false2023.12 | 76.7 | — | |
| AlignMamba-2Publication=PR’262026.03 | 73.1 | 71.5 | |
| AlignMamba-1Publication=CVPR’252026.03 | 72.4 | 70.9 | |
| EAUPublication=CVPR’242026.03 | 72.1 | 70.6 | |
| SigmaPublication=WACV’252026.03 | 71.8 | 70.3 | |
| EUARPublication=MM’242026.03 | 71.7 | 70.7 | |
| GeminiFusionPublication=ICML’242026.03 | 71.5 | 70 | |
| TMCPublication=ICLR’212026.03 | 71.1 | 69.8 | |
| TokenFusionPublication=CVPR’222026.03 | 70.8 | 69.5 | |
| QMFPublication=ICML’232026.03 | 70.1 | 68.7 | |
| CobraPublication=AAAI’252026.03 | 69.8 | 68.7 | |
| TRecgNet2022.01 | 69.2 | — | |
| Late FusionPublication=-2026.03 | 69.1 | 68.3 | |
| DF2Net2022.01 | 65.4 | — | |
| Ours2024.10 | 62.2 | — | |
| X-InstructBLIP2024.10 | 59.5 | — | |
| OneLLM2024.10 | 57.9 | — | |
| ImageBind0-shot=false2023.12 | 54 | — | |
| OneLLM-D0-shot=true, Input modality=Depth map2023.12 | 50.9 | — | |
| OneLLM-N0-shot=true, Input modality=Normal map2023.12 | 46.5 | — | |
| CLIP ViT-H*0-shot=true2023.12 | 41.9 | — | |
| Random2023.12 | 10 | — |