CurAItor | Devpost

Inspiration

CurAItor was born from our 🎨 love for art and the joy of visiting museums 🏛️—so we created it to bring a personalized museum experience to everyone, anywhere. As artists ourselves, we wanted to build an interactive, accessible space that empowers people to explore art in a fun and meaningful way.

Our mission is to:
✨ Connect people of all abilities with art that speaks to them
📚 Enhance art education and make learning about art more engaging
🌍 Open the door to a more inclusive and accessible art experience for everyone

With CurAItor, anyone can enjoy their own unique art gallery, anytime, anywhere! 🖼️💡

Product Summary

CurAItor is an AI-powered virtual art museum where users can enter a description or concept and explore a curated collection of real art pieces that bring their prompt to life.

🖼️ AI curates a personalized collection of artworks based on user input.
🏛️ Displays art in a 3D virtual museum that users can explore interactively.
🤣 Generates humorous descriptions for each piece to make art exploration fun and engaging.
🔊 Includes text-to-speech options for visually impaired users, making art more accessible to everyone.

Technology Stack and How We Built It

Front-End:

JavaScript and React: Built a responsive, user-friendly interface for users to input text prompts and explore curated art pieces.
Framer Motion: Used for animations and smooth transitions within the virtual museum.
Tailwind CSS: Styled the interface to make it clean, modern, and responsive.
3D Virtual Museum: Integrated an open-source GitHublink for an interactive, immersive experience.
Leonardo AI: Generated images used in the design of the homepage.

Back-End:

Python and Flask: Managed server-side logic and communication with AI models.
Cohere, Command-R, and OpenAI GPT-4: Processed user input to select artwork and generate artwork descriptions.
Text-to-Speech: Integrated OpenAI's GPT-4 for generating spoken descriptions for visually-impaired users.
API Calls: Facilitated communication between the front-end and back-end, ensuring real-time delivery of art pieces and descriptions.

Challenges we ran into

Front-End & UI: Styling the UI and positioning elements, especially placing the AI-generated paintings within the 3D room model, was complex.

Animation Issues: The door opening animation on Tailwind CSS took a lot of time to perfect.

Back-End Communication: Optimizing API calls for fast, reliable responses from AI models and ensuring correct artwork and realistic descriptions was challenging.

Integration: Merging the front-end and back-end to ensure smooth image display and API calls took time.

Despite these hurdles, we successfully refined the product! 🎉

Accomplishments that we're proud of

We’re thrilled with how our animations turned out, especially considering how long it took to perfect. As first-time users of AI tools, we’re excited about how we leveraged open API keys to integrate painting generation and text-to-speech descriptions. Despite challenges with Tailwind CSS and API communication, we overcame them and styled a sleek front-end. We’re proud to deliver a cohesive platform that greatly enhances art education and accessibility. This project has given us the confidence to dive into more advanced AI applications in the future.

What we learned

Through this project, we learned just how powerful AI can be when used in creative ways. Working with text-to-image generation models really showed us how much potential AI has in the art world. While AI is great at creating unique and impressive artwork, we quickly realized that getting it to accurately interpret user input in a creative way takes some fine-tuning. We also discovered how important it is to find a balance between the technical side of AI and making sure the app is still fast and easy for users to interact with. On the development side, we got a lot of hands-on experience with merging the front-end and back-end, especially in making sure React and Flask communicated smoothly. We learned just how important it is to optimize both the design and the server side so the app can handle requests in real-time and give users a smooth experience. From dealing with Tailwind CSS issues to tackling 3D rendering and API problems, we really got a deeper understanding of how tricky it can be to build a full-stack app, especially when combining AI with regular web development tools.

What's next for CurAItor

Celebrating women artists: Adding a feature that curates collections exclusively with pieces by women artists to highlight female contributions to art.
Enhanced AI models: Further improve the quality and variety of generated art by training the AI on a more diverse range of artistic styles and concepts.
User personalization: Implementing features like user profiles, where past interactions can inform future creations, could make the experience even more tailored.
Expanding the experience: Introduce more interactive features like VR integration for an immersive experience.