search

yassouali / CCT

:page_facing_up: Semi-Supervised Semantic Segmentation with Cross-Consistency Training (CVPR 2020).
https://yassouali.github.io/cct_page/
MIT License
395 stars 57 forks source link

Cityscapes/CamVid config file #49

Closed ghost closed 2 years ago

ghost commented 2 years ago

Hello,

I'm interested in running additional experiments using CCT on Cityscapes and possibly CamVid. Could you please provide a config.json for this setting similar to what was provided for VOC?

I expect to have to do some adjustments, but having the configuration you used would be much appreciated.

Thanks!

yassouali commented 2 years ago

Hi @blind-author

thank you for your intrest in our work

unfortunetely, i can't seem to find the version that was used for these experiments, however, to get the resutls, you simply need to adjust the dataloader of VOC for Cityscapes or CamVid and that it, from what i rember, the rest of the config are the same.

sorry for the delayed response

alittledou commented 1 year ago

Hello,

I'm interested in running additional experiments using CCT on Cityscapes and possibly CamVid. Could you please provide a config.json for this setting similar to what was provided for VOC?

I expect to have to do some adjustments, but having the configuration you used would be much appreciated.

Thanks!

Hello,have you tried the best Settings