Wavy Freeform (Progressive)
Like the Freeform model, the Wavy Freeform model generates a sequence of images that are amalgamations of single collection. Through training, these resulting images resolve more clearly and grow larger in resolution over time.
This model currently generates Playform’s largest images natively, without upscaling, at 1024x1024px. The resulting images have a characteristic “wavy” movement that differs depending on the quality of the collection used.
One collection of 30+ images is needed for the model, but a larger number, higher resolution, and more contextually similar images will create a better model.
Every 1k of iterations will generate 50 new Freeform images. The Progressive model starts off with 8x8px images, and slowly grows up to 1024x1024px with additional training.
From the My Projects screen, click Start New Project and choose the Progressive process.
Progressive models require one input set of images: Inspiration.
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 your model by clicking the yellow button on the upper right corner. The recommended training time is 3 hours, which roughly translates to 50k iterations and 2,500 result images.
Once your model has begun training, you can access your Results within just a few minutes. The first images begin relatively noisy and resolve more clearly with more training.
The more the better! The minimum is 30, but you can upload as many as 1000. The images should be at least 256 by 256 px in resolution. We accept PNGs and JPGs.
We find that a well-curated, large image set works really well. This image collection of Mark Zuckerberg photos is a good example. It has consistent cropping, and the eyes are "registered", meaning that in each image, the eyes are in the same place. However, the image collection would be better if it were higher resolution.
Note that you will get 50 result images per iteration, no matter how many images you used as input.
The results should look more and more similar to your input images as you train more iterations. However, after a few hundred iterations, any subsequent training will have very little effect on the results.