LDBE

Pytorch implementation for two papers (the paper will be released soon):

"Domain Adaptive Semantic Segmentation without Source Data", ACM MM2021.

"Challenging Source-free Domain Adaptive Semantic Segmentation", submitted to TPAMI.

Method

Result

GTA5 -> Cityscapes:

SYNTHIA -> Cityscapes:

Data

Download GTA5.

Download SYNTHIA. Please use SYNTHIA-RAND-CITYSCAPES

Download Cityscapes.

Make sure the data path is consistent with the path in config file.

Training (TODO)

Stage 0: Training on the source domain data.

Run "run_so.py". The trained model is available at ...

Stage 1: Label denoising (both positive learning and negative learning).

Set method:"ld" in config/ldbe_config.yml. Then, run "run.py". The trained model is available at ...

Stage 2: Boundary enhancement

Set method:"be" in config/ldbe_config.yml. Then, run "run.py". The trained model is available at ...

Acknowledgement

https://github.com/Solacex/CCM

https://github.com/yzou2/CRST

Contact

[email protected]