Panoptic Segmentation On Cityscapes Val
評価指標
AP
PQ
PQst
PQth
mIoU
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
比較表
| モデル名 | AP | PQ | PQst | PQth | mIoU |
|---|---|---|---|---|---|
| autofocusformer-image-segmentation-off-the | 46.2 | 67.7 | 71.5 | 62.5 | 83.0 |
| adaptis-adaptive-instance-selection-network | 36.3 | 62.0 | 64.4 | 58.7 | 79.2 |
| attention-guided-unified-network-for-panoptic | 34.4 | 59.0 | 62.1 | 54.8 | 75.6 |
| dilated-neighborhood-attention-transformer | 44.5 | 67.2 | - | - | 83.4 |
| learning-to-fuse-things-and-stuff | 39 | 60.4 | 63.3 | 56.1 | 78 |
| panoptic-deeplab-a-simple-strong-and-fast | 38.5 | 64.1 | - | - | 81.5 |
| adaptis-adaptive-instance-selection-network | 33.9 | 60.6 | 62.9 | 57.5 | 77.2 |
| fully-convolutional-networks-for-panoptic | - | - | 66.6 | - | - |
| cmt-deeplab-clustering-mask-transformers-for-1 | - | 64.6 | - | - | 81.4 |
| oneformer-one-transformer-to-rule-universal | 46.7 | 68.4 | - | - | 83.6 |
| weakly-and-semi-supervised-panoptic | 28.6 | 53.8 | 62.1 | 42.5 | 79.8 |
| deeperlab-single-shot-image-parser | - | 56.5 | - | - | - |
| axial-deeplab-stand-alone-axial-attention-for | 44.2 | 68.5 | - | - | 84.6 |
| combinatorial-optimization-for-panoptic | 34.1 | 62.1 | 67.2 | 55.1 | 79.3 |
| fully-convolutional-networks-for-panoptic | - | - | 70.6 | 59.5 | - |
| autofocusformer-image-segmentation-off-the | 44.2 | 66.9 | 70.8 | 61.5 | 82.2 |
| panoptic-feature-pyramid-networks | 33.0 | 58.1 | 62.5 | 52.0 | 75.7 |
| oneformer-one-transformer-to-rule-universal | 45.6 | 67.2 | - | - | 83.0 |
| learning-to-fuse-things-and-stuff | 37.6 | 59.2 | 61.5 | 56 | 77.8 |
| oneformer-one-transformer-to-rule-universal | 45.6 | 67.6 | - | - | 83.1 |
| oneformer-one-transformer-to-rule-universal | 48.7 | 70.1 | 74.1 | 64.6 | 84.6 |
| adaptis-adaptive-instance-selection-network | 32.3 | 59.0 | 61.3 | 55.8 | 75.3 |
| masked-attention-mask-transformer-for | 43.6 | 66.6 | - | - | 82.9 |
| efficientps-efficient-panoptic-segmentation | 39.1 | 64.9 | 67.7 | 61.0 | 90.3 |
| mask-r-cnn | - | - | - | 54.0 | - |
| intra-batch-supervision-for-panoptic-1 | - | 62.4 | 67.3 | 54.7 | - |
| k-means-mask-transformer | 44.0 | 68.4 | - | - | 83.5 |
| upsnet-a-unified-panoptic-segmentation | 37.8 | 60.5 | 63.0 | 57.0 | 77.8 |
| efficientps-efficient-panoptic-segmentation | 43.5 | 67.5 | 70.3 | 63.2 | 82.1 |
| scaling-wide-residual-networks-for-panoptic | 42.8 | 68.5 | - | - | 84.6 |
| upsnet-a-unified-panoptic-segmentation | 33.3 | 59.3 | 62.7 | 54.6 | 75.2 |
| scaling-wide-residual-networks-for-panoptic | 46.8 | 69.6 | - | - | 85.3 |
| fully-convolutional-networks-for-panoptic | - | 61.4 | - | 54.8 | - |
| panoptic-segmentation | 36.4 | 61.2 | 66.4 | 54 | - |
| oneformer-one-transformer-to-rule-universal | 46.5 | 68.51 | - | - | 83.0 |
| upsnet-a-unified-panoptic-segmentation | 39.0 | 61.8 | 64.8 | 57.6 | 79.2 |