Inspiration
I regularly visit art museums, but I often leave with the feeling that I am not enjoying the experience to the fullest due to the lack of knowledge of art. In one of the anime I've recently watched, Blue Period, a character said, "to me, art is something like uneatable food. It's natural to have likes and dislikes. Expensive dishes don't always taste good. Knowing the place and the method of production, I might think it's delicious." From the combination of this idea and the recent hype of ChatGPT, I decided to build an app that can enrich our art museum experience with the power of machine learning.
What it does
MuseumMuse is a mobile app that you can take to art museums and instantly get more information about artworks by scanning their labels.
How we built it
We built the app with React Native in TypeScript and leveraged the OpenAI API to complete the sentence requested with the scanned text.
Challenges we ran into
- Time constraints
- Coming up with straightforward UI/UX
- Adjusting parameters for the sentence completion to get a relatively reliable response
Accomplishments that we're proud of
- Utilized AI technology to solve real-world problems
- Created a fully functional app with the expected features
What we learned
- AI is powerful, and it largely depends on us not only to create reliable and scalable models but also to find and implement their applications to solve problems in real life
What's next for MuseumMuse
- Bookmark feature
- Implement a recommendation engine to suggest artists or artworks you might like
- Audio support for accessibility
- Internationalization
- Android support (to be tested)
- Add authentication and subscription model
- More features (TBD) to make the art museum experience better
Built With
- openai
- react-native
- tailwind
- typescript

Log in or sign up for Devpost to join the conversation.