Starting off with some background

I'm working on a passion project for a group of people known as furries. The Furry Fandom in general are people who are interested in animals with human characteristics. You can read more about the fandom in general at Wikipedia.

Furries hold many conventions a year. Costumes known as fursuits are common, though the majority of furries do not own one. Many furries only relation with fursuits is to take pictures.

After the convention, often photographers want to identify the fursuit in their photos, to accurately describe them. (Many fursuiters do not speak in fursuit, so photographers often don't get the fursuiter's name.) Fursuiters often search social media after a convention, since that's their memory of their time in fursuit.

I'm looking for a way to bring the two together.

Machine Learning has advanced a lot in the past few years. Given a set of correctly labelled images, a computer can pick out relevant details, and create a model of what is relevant to correctly identify an image of the same thing in the future.

So that's my plan. Take a bunch of correctly labelled images of fursuits, and create a Machine Learning model that can name the fursuit in any new photo.

• Photographers can upload photos with fursuits they don't know. The model would instantly reply back with the name of the fursuit.
• Fursuiters may get a benefit out of this even if they don't hear about this service. More photographers will be able to identify the fursuits in their photos, and credit the fursuiter when they post to social media.