stelledge / neural-3d-face Goto Github PK

I am interested in using the Neural-3D-Face model for a project, and I would like to train it on my own dataset. I have reviewed the documentation and issues but haven't found specific guidelines on how to adapt the training process for a custom dataset.

Could you please provide detailed steps or guidelines on how to prepare and integrate a custom dataset into the training pipeline? Additionally, are there any specific requirements or modifications that need to be made to the dataset format or the training code?

I am particularly interested in understanding the following:

• The format and structure required for the dataset (e.g., file types, annotations, directory structure).
• Any preprocessing steps that are necessary before training.
• Adjustments to the training parameters or configuration files that might be required for a custom dataset.

Thank you very much for your help and for developing this interesting project!