I’m working on an API that has very unique image processing effects using deep learning. So the ability to take pictures and paint them in different styles or hallucinate objects into them: http://www.somatic.io/models/popular
Would something like this be interesting to you all?
It would be a non free API because the servers use expensive GPUs. If there is real interest in this, then I will port it to vvvv.
Anyone, please feel free to contact me.
To be honest I didn’t find the output of your models particularly unique so I don’t think I would personally be interested in using your service for that. It’s a technique that’s becoming popular to the point of kitch: it will probably be an ironic internet fad within the next 6 months, like face-swapping and “I am T-Pain” apps. Cash in while you can I suppose.
However, people may be interested in using it for now and other parts of your api, and since people have to get an api-key to use your service I’d say give to a whirl. It can’t hurt to make another option available to use your product, especially a creative coding environment such as vvvv.
My goal is to make all the deep learning models for image processing available to everyone. Right now its very painful to get these all up and running. The list pf api end points/models would keep growing overtime.