Panoptic Segmentation On Ade20K Val
Metriken
AP
PQ
mIoU
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Vergleichstabelle
| Modellname | AP | PQ | mIoU |
|---|---|---|---|
| masked-attention-mask-transformer-for | 34.2 | 48.1 | 54.5 |
| oneformer-one-transformer-to-rule-universal | 36.2 | 50.0 | 56.6 |
| a-simple-framework-for-open-vocabulary | - | 53.7 | - |
| oneformer-one-transformer-to-rule-universal | 36.0 | 50.5 | 58.3 |
| generalized-decoding-for-pixel-image-and | 38.7 | 52.4 | 59.1 |
| dilated-neighborhood-attention-transformer | 35.0 | 49.4 | 56.3 |
| generalized-decoding-for-pixel-image-and | 35.8 | 49.6 | 58.1 |
| masked-attention-mask-transformer-for | 26.5 | - | 46.1 |
| masked-attention-mask-transformer-for | - | 39.7 | - |
| k-means-mask-transformer | - | 42.3 | 45.3 |
| k-means-mask-transformer | - | 50.9 | 55.2 |
| masked-attention-mask-transformer-for | 33.2 | 46.2 | 55.4 |
| k-means-mask-transformer | - | 48.7 | 54.8 |
| oneformer-one-transformer-to-rule-universal | 40.2 | 54.5 | 60.4 |
| oneformer-one-transformer-to-rule-universal | - | 53.4 | 58.9 |
| per-pixel-classification-is-not-all-you-need | - | 35.7 | - |
| oneformer-one-transformer-to-rule-universal | 36.3 | 50.1 | 57.4 |
| oneformer-one-transformer-to-rule-universal | 37.1 | 51.5 | 58.3 |
| oneformer-one-transformer-to-rule-universal | 37.8 | 51.4 | 57.0 |
| k-means-mask-transformer | - | 41.5 | 45.0 |
| oneformer-one-transformer-to-rule-universal | 35.9 | 49.8 | 57.0 |
| masked-attention-mask-transformer-for | - | 37.9 | 50 |