training
Opened this issue · 3 comments
Hello, I would like to ask if you are using the complete MegaDepth dataset for training data, or select a part of it, and if it is convenient, can you provide a training data?
As stated in https://github.com/ubc-vision/COTR/blob/master/prepare_data.md, we use scenes [0000, 0240] EXCEPT scene 0204 as training split, we use scene 0204 as the validation split, and the rest as the test split.
You can check this markdown file, we detailed the training process. We also provided the download of the distance matrix for all the scenes in Megadepth.
Hello, very thanks for your project, but the raw MegaDepth dataset is too larger to download, could you please share the dataset link used in the project？ thanks very much！
Sorry, we are not able to share/host the subset, can't help with that...