Details of MegaDepth for WarpC training.

PruneTruong / DenseMatching

Dense matching library based on PyTorch

GNU Lesser General Public License v2.1

689 stars 81 forks source link

Closed wdmwhh closed 2 years ago

wdmwhh commented 2 years ago

Thanks for your great works. I fail to find

'train_split': 'train_scenes_MegaDepth.txt',
'train_debug_split': 'train_debug_scenes_MegaDepth.txt',
'val_split': 'validation_scenes_MegaDepth.txt',

these files. Could you release more details to preprocess MegaDepth Dataset?

PruneTruong commented 2 years ago

Hi, the files are in the datasets/MegaDepth folder (https://github.com/PruneTruong/DenseMatching/tree/main/datasets/MegaDepth). The train_debug_split is just one scene is use to debug on my computer. You can train WarpC-GLUNet by running this file (https://github.com/PruneTruong/DenseMatching/blob/main/train_settings/WarpC/train_WarpC_GLUNet_stage1.py).

wdmwhh commented 2 years ago

Thanks for your guidance. I am careless.