Train with my own dataset

lliuz / ARFlow

The official PyTorch implementation of the paper "Learning by Analogy: Reliable Supervision from Transformations for Unsupervised Optical Flow Estimation".

MIT License

249 stars 49 forks source link

Train with my own dataset #37

Open sotomotocross opened 3 years ago

sotomotocross commented 3 years ago

If i have my own unlabeled dataset (no ground truth) on which i want to train the ARFlow network. How can i do it? Is there a rule of thumb of organizing the data in order to train the NN?

zerotwotnt commented 2 years ago

I have the same questions ,How can I train my own dataset.

shinaushin commented 1 year ago

I have same question too

mmitchef commented 6 months ago

Hi @shinaushin @zerotwotnt @sotomotocross Could any of you figure out how to use this on custom dataset without label?