Which pictures of Cityscapes dataset are use to make _qualitative results of semantic segmentation_?

microsoft / ProDA

Prototypical Pseudo Label Denoising and Target Structure Learning for Domain Adaptive Semantic Segmentation (CVPR 2021)

https://arxiv.org/abs/2101.10979

MIT License

286 stars 44 forks source link

Which pictures of Cityscapes dataset are use to make _qualitative results of semantic segmentation_? #51

Open ZHE-SAPI opened 2 years ago

ZHE-SAPI commented 2 years ago

Good job!! I would like to ask which pictures of Cityscapes dataset are use to make qualitative results of semantic segmentation, 500 val images? thanks!

ZHE-SAPI commented 2 years ago

@zhangmozhe @msftdata @msftgits @dnfclas Could you please answer my question? thx!