| OneFormer: One Transformer to Rule Universal Image Segmentation | ✓ Link | 46.7 | 61.7 | 54.9 | 40.5 | OneFormer (DiNAT-L, single-scale) | 2022-11-10 |
| OneFormer: One Transformer to Rule Universal Image Segmentation | ✓ Link | 46.4 | 61.6 | 54.0 | 40.6 | OneFormer (ConvNeXt-L, single-scale) | 2022-11-10 |
| Fully Convolutional Networks for Panoptic Segmentation | ✓ Link | 45.7 | | 52.1 | 40.8 | Panoptic FCN* (Swin-L, single-scale) | 2020-12-01 |
| Scaling Wide Residual Networks for Panoptic Segmentation | | 44.8 | 60.0 | 51.9 | 39.3 | Panoptic-DeepLab (SWideRNet-(1, 1, 4.5), multi-scale) | 2020-11-23 |
| Intra-Batch Supervision for Panoptic Segmentation on High-Resolution Images | ✓ Link | 42.2 | | 52.0 | 34.9 | Mask2Former + Intra-Batch Supervision (ResNet-50) | 2023-04-17 |
| Axial-DeepLab: Stand-Alone Axial-Attention for Panoptic Segmentation | ✓ Link | 41.1 | 58.4 | 51.3 | 33.4 | Axial-DeepLab-L (multi-scale) | 2020-03-17 |
| EfficientPS: Efficient Panoptic Segmentation | ✓ Link | 40.6 | | | | EfficientPS | 2020-04-05 |
| Panoptic-DeepLab: A Simple, Strong, and Fast Baseline for Bottom-Up Panoptic Segmentation | ✓ Link | 40.5 | | | | Panoptic-DeepLab (X71) | 2019-11-22 |
| AdaptIS: Adaptive Instance Selection Network | | 40.3 | 56.8 | | | AdaptIS (ResNeXt-101) | 2019-09-17 |
| Fully Convolutional Networks for Panoptic Segmentation | ✓ Link | 36.9 | | | 32.9 | Panoptic FCN* (ResNet-FPN) | 2020-12-01 |
| Panoptic Segmentation with a Joint Semantic and Instance Segmentation Network | | 17.6 | | | | JSIS-Net (ResNet-50) | 2018-09-06 |
| Hierarchical Multi-Scale Attention for Semantic Segmentation | ✓ Link | 17.6 | | | | HRNet-OCR (Hierarchical Multi-Scale Attention) | 2020-05-21 |
| Fully Convolutional Networks for Panoptic Segmentation | ✓ Link | | | 42.3 | | Panoptic FCN* (ResNet-50-FPN) | 2020-12-01 |