SnapSphere

Discover hidden photo spots with vibe-based maps, create personalized moodboards, explore community-shared locations, and enjoy AI vibe classification — making every shoot uniquely yours.

Comment

Inspiration

The idea for Snapsphere came from a simple frustration: traditional maps and location apps often feel cold and uninspiring. As someone who loves photography and exploring new places, I struggled to find spots that truly matched the mood or aesthetic I wanted—places often overlooked by popular guides. I realized what makes a location special isn’t just popularity —it’s its vibe: the light, colors, textures, and feeling it evokes. From misty alleys to sunlit parks, these qualities shape how we experience and capture places. Inspired by this gap, I imagined a tool blending map precision with the creativity of platforms like Pinterest. What if discovering locations was less about pins and ratings, and more about feeling? What if AI could understand mood and guide you to spots that resonate with your style? Snapsphere was born as a digital compass for photographers, content creators, and anyone eager to explore vibe-rich places or capture unique photos. Combining AI image analysis, geospatial data, and community input, it transforms how we explore and capture the world, making every shoot uniquely inspired.

What it does

Snapsphere uses Google maps, AI and community tips to help photographers and creators find the perfect spots that match their style and the moment. Its key features include:

  • VibeMap: An interactive map displaying photo spots enriched with AI-analyzed vibes and aesthetic classifications.
  • Moodboards: A visual storyboarding tool to plan shoots by assembling photos and locations into mood-driven boards.
  • Community Photo Spots: User-submitted, verified photo locations with timing tips, safety notes, and sample shots.
  • AI Vibe Detection: Upload photos to receive AI-powered analysis of vibe, color palettes, and style characteristics.
  • Smart Recommendations: Personalized suggestions based on weather, time of day, and user style preferences.
  • Caption Generator: AI-crafted Instagram-worthy captions with relevant hashtags tailored to the photo’s mood and location.

How we built it

We built SnapSphere using a modern full-stack architecture with Python Flask backend and React TypeScript frontend. The core AI functionality leverages OpenAI's CLIP model through PyTorch and Transformers for real-time image vibe classification and aesthetic analysis. We integrated Mapillary's street-level imagery API for location-based photo discovery, combined with Nominatim reverse geocoding to convert coordinates into human-readable addresses. The application uses SQLite database with SQLAlchemy ORM for data persistence, while the frontend employs Tailwind CSS for responsive design and Google Maps API for interactive location visualization. The system processes base64-encoded images through CLIP for multi-label classification, generating intelligent captions and mood-based recommendations using custom prompt engineering and confidence scoring algorithms.

Challenges we ran into

One of the main challenges was integrating Google Maps features like reverse geocoding and Mapillary imagery to display photo spots filtered by vibe. Matching AI-generated aesthetics with accurate map data required careful coordination. We also faced hurdles with image handling and ensuring that the CLIP model could analyze them in real time without slowing things down. Port mismatches between frontend and backend, optimizing the database for user preferences, and balancing secure login with a smooth demo experience were other key technical challenges.

Accomplishments that we're proud of

We're proud of building a fully functional MVP that combines AI, geospatial data, and community input into a cohesive experience. We successfully integrated Google Maps with reverse geocoding, used Mapillary for visual context, and deployed an AI model that classifies image vibes in real time. Creating features like vibemap, moodboard generation, and smart recommendations in a short time frame was both exciting and rewarding.

What we learned

We learned a lot about combining AI with geospatial tools in a creative, user-centric way. From handling different image formats to deploying CLIP and working with Google Maps APIs, we gained hands-on experience in managing both frontend and backend complexities. We also realized how crucial it is to fine-tune performance when using AI in real time — even small delays can affect the overall feel. Beyond the tech, we learned how to keep our design grounded in the user’s journey: making user interaction inspiring rather than overwhelming.

What's next for SnapSphere

We’re excited to keep building on what we’ve started! Coming soon are more detailed vibe categories (like “sunset nostalgia” or “coastal calm”), smarter AI captions for social sharing, and moodboards that are fully customizable and collaborative. We also plan to roll out Golden Hour alerts so users can catch the best lighting for their shoots, and introduce safe route suggestions to help users explore new areas confidently. A mobile version is in the works to make Snapsphere more accessible on the go. Long-term, we hope to grow Snapsphere into a vibrant creative community — where photographers and explorers can follow each other, create vibe trails, and share their favorite hidden gems with ease.

Built With

Share this project:

Updates