Panoptic Segmentation On Coco Test Dev
Metriken
PQ
PQst
PQth
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Vergleichstabelle
| Modellname | PQ | PQst | PQth |
|---|---|---|---|
| attention-guided-unified-network-for-panoptic | 45.2 | 31.3 | 54.4 |
| refine-prediction-fusion-network-for-panoptic | 49.6 | 37.7 | 57.5 |
| cmt-deeplab-clustering-mask-transformers-for-1 | 55.7 | 46.8 | 61.6 |
| axial-deeplab-stand-alone-axial-attention-for | 44.2 | 36.8 | 49.2 |
| masked-attention-mask-transformer-for | 58.3 | 48.1 | 65.1 |
| panoptic-deeplab-a-simple-strong-and-fast | 41.4 | 35.9 | 45.1 |
| adaptis-adaptive-instance-selection-network | 42.8 | 31.8 | 50.1 |
| scaling-wide-residual-networks-for-panoptic | 46.5 | 38.2 | 52.0 |
| k-net-towards-unified-image-segmentation | 55.2 | 46.2 | 61.2 |
| refine-prediction-fusion-network-for-panoptic | 51.5 | 39.2 | 59.6 |
| panoptic-segformer | 50.9 | 43.0 | 56.2 |
| panoptic-segformer | 56.2 | 47.0 | 62.3 |
| panoptic-segmentation-with-a-joint-semantic | 27.2 | 23.4 | 29.6 |
| max-deeplab-end-to-end-panoptic-segmentation | 51.3 | 42.4 | 57.2 |
| attention-guided-unified-network-for-panoptic | 45.5 | 31.6 | 54.7 |
| learning-instance-occlusion-for-panoptic | 46.6 | 35.7 | 54.0 |
| combinatorial-optimization-for-panoptic | 38.5 | 34.8 | 41.0 |
| k-net-towards-unified-image-segmentation | 48.3 | 39.7 | 54 |
| fully-convolutional-networks-for-panoptic | 47.5 | 38.2 | 53.7 |
| maskconver-revisiting-pure-convolution-model | 53.6 | 58.9 | 45.6 |
| panoptic-feature-pyramid-networks | 40.9 | 29.7 | 48.3 |
| axial-deeplab-stand-alone-axial-attention-for | 43.6 | 35.6 | 48.9 |
| epsnet-efficient-panoptic-segmentation | 38.9 | 31.0 | 44.1 |
| k-means-mask-transformer | 58.5 | 49.0 | 64.8 |
| sognet-scene-overlap-graph-network-for | 47.8 | - | - |
| mask-dino-towards-a-unified-transformer-based-1 | 59.5 | - | - |
| learning-to-fuse-things-and-stuff | 40.7 | 31.0 | 47.0 |
| generator-evaluator-selector-net-a-modular | 33.7 | 31.5 | 35.1 |
| panoptic-segformer | 55.8 | 46.5 | 61.9 |
| ada-segment-automated-multi-loss-adaptation | 48.5 | 37.6 | 55.7 |
| per-pixel-classification-is-not-all-you-need | 53.3 | 44.5 | 59.1 |
| detectors-detecting-objects-with-recursive-1 | 50 | 37.2 | 58.5 |
| upsnet-a-unified-panoptic-segmentation | 46.6 | 36.7 | 53.2 |
| pixel-consensus-voting-for-panoptic | 37.7 | 33.1 | 40.7 |
| attention-guided-unified-network-for-panoptic | 46.5 | 32.5 | 55.8 |
| fully-convolutional-networks-for-panoptic | 52.7 | - | 59.4 |
| panoptic-segformer | 50.2 | 42.4 | 55.3 |
| spatialflow-bridging-all-tasks-for-panoptic | 48.5 | 37.9 | 55.5 |