MIC
MIC copied to clipboard
lhoyer
[CVPR23] Official Implementation of MIC: Masked Image Consistency for Context-Enhanced Domain Adaptation
Hi. I have a question about inference. As I understand, teacher model is more robust than student model by EMA update and it is used for supervision(pseudo label). Then why...
I have some problems with the data set used in the classification section. In the classification experiment, is only the training set and validation set of the VisDA-2017 data set...
The given config file uses dacs as uda, does anyone know how to construct the config file for advseg? Thank you in advance.
Before asking my question, thank you for sharing your work! I found that color augmentation is deterministic in your classification code. In classification code(cls/dalib/modules/masking.py line 100-109), self.augmentation_params['color_jitter'] and self.augmentation_params['blur'] is...
I encountered a weired issue that same basic config (GTA->CItyscapes:crop size1024x2048) took up different GPU memory on RTX3090(total 24G) with 23.34G and RTX A6000(total 48G) with 40G。
Is it possible to use any other loss function instead of cross-entropy? I found where the cross-entropy loss function has been defined. However I did not find where it is...
Dear Lucas, I encountered a significant performance variance while training DAFormer with MIC. The lowest performance reached was only 64.57, which is far below the reported 70.6 in the paper....
Dear author, Thank you so much for such a great job! I have encountered a problem in my experiment and hope to get your guidance. In the experiment, we found...
 Hello. I trained your model on my custom dataset and its target data size is 2048x400 (width, height). When I train it with above config (resize 2048x400), its performance...
Dear authors, I'm trying to reproduce the office-home experiments in the `cls` directory with `python run_experiments.py --exp 1`. However, I commented out Line 117 `('VisDA2017', 'Synthetic', 'Real')`. When running the...
