Open Vocabulary Panoptic Segmentation On
Metrics
PQ
Results
Performance results of various models on this benchmark
Model Name | PQ | Paper Title | Repository |
|---|---|---|---|
| PosSAM | 29.2 | PosSAM: Panoptic Open-vocabulary Segment Anything | - |
| ODISE (Label) | 22.6 | Open-Vocabulary Panoptic Segmentation with Text-to-Image Diffusion Models | - |
| ODISE(Caption) | 23.4 | Open-Vocabulary Panoptic Segmentation with Text-to-Image Diffusion Models | - |
| MAFT+ | 27.1 | Collaborative Vision-Text Representation Optimizing for Open-Vocabulary Segmentation | - |
| MaskCLIP | 15.1 | Extract Free Dense Labels from CLIP | - |
| CLIPSelf | 23.7 | CLIPSelf: Vision Transformer Distills Itself for Open-Vocabulary Dense Prediction | - |
| FreeSeg | 16.3 | FreeSeg: Unified, Universal and Open-Vocabulary Image Segmentation | - |
| FC-CLIP | 26.8 | Convolutions Die Hard: Open-Vocabulary Segmentation with Single Frozen Convolutional CLIP | - |
0 of 8 row(s) selected.