Creative Morph
Last updated
Last updated
This process generates a sequence of images that initially resemble one set of input images and then gradually converge toward a second set. The images in-between are exploratory mixtures of the two.
The Creative Morph process requires two sets of input images, called Inspiration and Influences, with 30+ images in each. As the model learns how to translate from one image set to the other, it attempts to preserve the contours and shapes present in the Inspiration images, while taking cues from the colors within the Influence images.
Each iteration of training generates one output image per Inspiration image provided. For instance, 200 Inspiration images trained for 50 snapshots will create 10,000 result images (200 per iteration).
Currently, this model outputs 512 x 512 px results that can be upscaled.
From the Processes tab under Train a Model, choose Creative Morph, then click Create Project.
The Creative Morph model requires at least two collections of images: one in Inspiration and another in Influences. You can also upload multiple collections for each input set.
You will need to Upload Images to make a new collection or choose a Recently Used collection.
Only selected images will be used in the model.
You can start training by clicking the yellow button on the upper right corner. The recommended training time is 1 hour and 30 minutes, which roughly translates to 50 snapshots.
Once your model has begun training, you can access your results within just a few minutes. These first images will start off looking like the Inspiration images and slowly converge towards the Influence images.
Traditionally, it takes exponentially longer to train a model to create larger results. Because of this, we have optimized the process to create 512px images for you to be able to experiment more quickly. You can also upscale your images in the single image view, with more sophisticated AI upscale methods available for Plus members.
This will not challenge the Creative Morph process, so you will end up getting the exact same images you inputted. Since you do not want to be spending resources to get no results, we will make sure to alert you if you have redundant collections.
