yuhuayc/da-faster-rcnn

Details of source target setting for KITTI to Cityscapes and Cityscapes to KITTI

Closed this issue · 1 comments

@yuhuayc Hi! yuhua, thanks for your wonderful work, I'm wondering for the K->C, do you use KITTI full training data as source, and Cityscapes train as target unlabeled data, Cityscapes val as target test data? for C->K, do you use Cityscapes full trainval as source training data, and split KITTI for target unlabeled data and test data? I'm reproducing this part and haven't found details about this in code and paper, looking forward to you reply!

Thanks for your message.

For K->C, kitti full training data is used as as source, and Cityscapes train as target unlabeled data, Cityscapes val as target test data.

For C->K, Cityscapes train is used as source, and Kitti is used both as target unlabeled data and target test data without split.