Image Display Control - The next generation of responsive images

@kallefreese super interesting question, and something I think about on a daily basis. So far, cropping and composition have not been given serious consideration in these models. Much of the current training data is based on center-cropped images, resulting in monotonous results. But even if ML models would be capable of taking the limitations and restrictions of the “frame” better into account, that still only solves part of the problem. The image will still need to be shown in different sizes and aspect ratios on different devices and platforms. Without Image Display Control, it cannot adapt correctly to all those containers. So Frameright and IDC complement also AI-generated imagery. And to make the process smoother, our own Frameright AI can be used to provide crop composition suggestions. This is not the same thing as generative models creating multiple versions of the same image as hard-crops. Instead, we aim to enable one image asset to adapt to multiple shapes and sizes. My motivation for building IDC is to ensure that there are the right tools for rich visual storytelling in the future for both machine-generated and human-generated content. As I am writing this, my mind is filled with new possibilities for how images could work. There are many developments that impact and support our business. We are excited to be at the forefront of these advancements.

@gerhard_raab excatly 🎯 Thank you so much for the comment 🙏

@felipe_millan Thank you, let's make that happen!

@jimmy_mailer Thank you 🙏

@toniperamaki Thank you for your kind words. It has been a pleasure getting to know you ✨