On this page you can find code that will allow you to explore different 2D-to-2D and 2D-to-3D image editing techniques, as well as collect simple datasets. The majority of these editing techniques must be run on graphics processing units (GPUs) in order to run or train efficiently and in a reasonable time frame. Thus, this repository is designed to run on Paperspace, which is a cloud computing platform that allows you to access GPUs. Tutorials on how to use the code in this repository with Paperspace is given in the following section.
For step by step instructions on using Paperspace, please see the following:
Uploading your dataset to and downloading your model outputs from Paperspace
Running and training models using the Gradient Experiment Builder
To access the tutorials for how to run/train a variety of different models using the Paperspace Gradient Experiment Builder:
For a step-by-step walk through of the above folders and files and the code they contain: