Open Vocabulary Semantic Segmentation On 2

mIoU

评测结果

各个模型在此基准测试上的表现结果

比较表格

模型名称	mIoU
ttd-text-tag-self-distillation-enhancing	12.7
learning-mask-aware-clip-representations-for	32.0
convolutions-die-hard-open-vocabulary-1	34.1
collaborative-vision-text-representation	36.1
cat-seg-cost-aggregation-for-open-vocabulary	37.9
opendas-domain-adaptation-for-open-vocabulary	35.8
2112-14757	20.5
open-vocabulary-segmentation-with-semantic	33.5
mask-adapter-the-devil-is-in-the-masks-for	38.2
open-vocabulary-panoptic-segmentation-with	23.7
模型 11	20.7
open-vocabulary-semantic-segmentation-with-4	32.8
ttd-text-tag-self-distillation-enhancing	17.0
open-vocabulary-panoptic-segmentation-with-1	29.9
open-vocabulary-semantic-segmentation-with-2	31.4
silc-improving-vision-language-pretraining	37.7
open-vocabulary-semantic-segmentation-with	29.6
in-defense-of-lazy-visual-grounding-for-open	15.8
maskclip-a-mask-based-clip-fine-tuning	38.2
clipself-vision-transformer-distills-itself	34.5
sed-a-simple-encoder-decoder-for-open	35.2