Well, time passed and i finally got to the github, so now i can complete some people's request to publish my solution on parsing google's best guess html code without the help of the external server.
Also there are two significant editions that i made:
- I completely re-wrote the original code in order to structurize it and make it more readable.
- Added the Microsoft Vision API support for the image recognition - now a user can choose which service to use (Google or Microsoft).
- If no info was gotten from the wikipedia (for the description of the recognized object) the app tries to get some from the Oxford Dictionary API.
You'll have to obtain your own API keys for the following services:
The process of obtaining the first two keys was kindly recorded by Matthew in his original video.