cnn-facial-landmark
cnn-facial-landmark copied to clipboard
landmark.py File structure
Hello Moderator,can you tell me what the file structure is during training? Or when running the landmark.py file, in which folder are the three .record files in the root directory? What documents are needed for training? Also, how to print the test effect and loss value after training?
Hello , Take a look on this tutorial : https://github.com/faust690226/cnn-facial-landmark-tutorial
Does this mean that I can collect some facial photos and building a dataset for training from these? I am having a hard time to understand how to do the training.
Training with custom dataset is possible.
The training should not be hard as most of the training code is already available. The datasets ready for training however, need a lot of works to do.