I find there are many four-grid images (e.g. link) generated by midjourney in your image dataset, such as Midjourney-v5-1.7M.
How do you apply these four-grid images in your training stage? Using directly or splitting a four-grid image into four ones
I find there are many four-grid images (e.g. link) generated by midjourney in your image dataset, such as Midjourney-v5-1.7M. How do you apply these four-grid images in your training stage? Using directly or splitting a four-grid image into four ones