Unsupervised Semantic Segmentation With 8
المقاييس
mIoU
النتائج
نتائج أداء النماذج المختلفة على هذا المعيار القياسي
جدول المقارنة
| اسم النموذج | mIoU |
|---|---|
| denseclip-extract-free-dense-labels-from-clip | 26.4 |
| ttd-text-tag-self-distillation-enhancing | 31.0 |
| tagalign-improving-vision-language-alignment | 37.6 |
| learning-to-generate-text-grounded-mask-for | 33.9 |
| reco-retrieve-and-co-segment-for-zero-shot-1 | 22.3 |
| proxyclip-proxy-attention-improves-clip-for | 39.6 |
| ttd-text-tag-self-distillation-enhancing | 37.4 |
| cosmos-cross-modality-self-distillation-for | 33.7 |
| groupvit-semantic-segmentation-emerges-from | 23.4 |
| harnessing-vision-foundation-models-for-high | 44.3 |