What it does
ArtID takes an image file as user input and uses Google Cloud Vision to search the web for its page on the Metropolitan Museum's website. It gets the unique identification number of the artwork from the end of the URL and then uses this ID to search a JSON file for information such as title, artist, time period, and medium, as well as retrieves a text file that uses the ID as its name and contains the artwork's catalogue entry. All of this information about the artwork is then displayed for the user.
How I built it
The program is built using Node.js, using Google Cloud Vision for image identification and JSON for data storage.
Challenges I ran into
I attempted to use Google Cloud SQL for data storage, but had trouble connecting to the database in Node.js. I also had trouble figuring out how to connect a frontend to Node.js.
What I learned
How to use Node.js and Google Cloud Vision, neither of which I have ever used before.
What's next for ArtID
Taking photos of artwork to identify instead of just uploading an image. A more comprehensive database that goes beyond the Met's collection.
Built With
- google-cloud
- google-cloud-vision
- json
- node.js
Log in or sign up for Devpost to join the conversation.