Free-Form-Video-Inpainting
Free-Form-Video-Inpainting copied to clipboard
About the training dataset
Thanks for sharing your work. but,when I use the FVI dataset, root_masks_dir="../dataset/FVI/Train/object_masks/" Displays: NotImplementedError: Image mode P And, after using root_masks_dir="../dataset/FVI/Train/random_masks/" The number of JPEGImages trained in the epoch is 1940, and the random_masks are greater than 2,000, so a complete epoch cannot be completed.
@caimaomao315 I have the same problem. Have you solved it?
@caimaomao315
The number of folders under random_masks
is 24000. I will try to delete some folders in random_masks
to make it 1940 and see what will happen.
//Edited later//
I modified save_freq
in src/config.json
to 1 (instead of 5), and it seems that it can save the checkpoint file smoothly at the end of each epoch.