Icon Recognition Software
I’m looking for a team to develop Icon Recognition Software specifically to help app developers search our database of app icons to see if their app icons are original, or to similar to other app icons.
Our database has approximately 2 million app icons from IOS, Android, Microsoft, Xbox, Facebook, etc.
We’re looking for someone with a background in computer vision to program an algorithm to compare and detect similarities between two icon or logo images.
An example of an approach to achieve this can be seen here, in particular the ‘final phase’ of the pipeline: http://dl.acm.org/citation.cfm?id=959705
We are open to any reasonable approaches to achieving this functionality and to using existing code or techniques.
The final product will be in the form of a website and the image processing will be done server-side.
Our database (MongoDB) is running on Heroku. If you are able to set up the server-side (whether OpenCV or otherwise) environment and API for our front-end developer to communicate with that would be a plus!
Please see attached.
Please ask me questions