Open-vocabulary semantic segmentation on PASCAL VOC PAS-20
95mIoUHyRo
Evaluation results
| Method | Links | |
|---|---|---|
| HyRoVLM=CLIP ViT-B/16, Fine-tuning Space=H2026.05 | 95 | |
| HyperCLIPVLM=CLIP ViT-B/16, Fine-tuning Space=H2026.05 | 94.9 | |
| SEDVLM=CLIP ConvNeXt-B, Fine-tuning Space=E2026.05 | 94.4 | |
| SANVLM=CLIP ViT-B/16, Additional Backbone=Side Adapter, Fine-tuning Space=E2026.05 | 94 | |
| ZegCLIPVLM=CLIP ViT-B/16, Fine-tuning Space=E2026.05 | 93.6 | |
| OVSegVLM=CLIP ViT-B/16, Additional Backbone=ResNet-101c, Fine-tuning Space=E2026.05 | 92.6 | |
| ZSsegVLM=CLIP ViT-B/16, Additional Backbone=ResNet-101, Fine-tuning Space=E2026.05 | 88.4 | |
| ZegFormerVLM=CLIP ViT-B/16, Additional Backbone=ResNet-101, Fine-tuning Space=E2026.05 | 86.2 | |
| LSegVLM=CLIP ViT-B/32, Additional Backbone=ResNet-101, Fine-tuning Space=E2026.05 | 47.4 | |
| ZS3NetAdditional Backbone=ResNet-101, Fine-tuning Space=E2026.05 | 38.3 |