| Benchmark | Model | Rank | Results |
|---|---|---|---|
| open-vocabulary-object-detection-on-lvis-v1-0 | CLIPSelf | #5 | AP novel-LVIS base training: 34.9 |
| open-vocabulary-object-detection-on-mscoco | CLIPSelf | #6 | AP 0.5: 44.3 |
| open-vocabulary-panoptic-segmentation-on-ade20k | CLIPSelf | #6 | PQ: 23.7 |
| open-vocabulary-semantic-segmentation-on-1 | CLIPSelf | #4 | mIoU: 62.3 |
| open-vocabulary-semantic-segmentation-on-2 | CLIPSelf | #8 | mIoU: 34.5 |
| open-vocabulary-semantic-segmentation-on-3 | CLIPSelf | #13 | mIoU: 12.4 |