Find My Bookv1.0
✦ Project Overview
An AI-powered personal librarian that recommends books based on visual input. Users can upload a photo of a bookshelf, and the system (powered by Google Gemini) identifies books and suggests the best reads based on user mood and preferences.
✦ Key Features
- ♥Visual book recognition from bookshelf photos
- ♥Mood-based book recommendations via Gemini AI
- ♥Personalized suggestions based on reading history
- ♥Simple and intuitive user interface
✦ Methodology
Integrates Multimodal AI for visual understanding:
01.
Image Processing
User uploaded images are passed to the Gemini Vision Pro model, which extracts titles and authors from the book spines.
02.
Contextual Matching
The extracted book list is cross-referenced with the user's stated mood/preferences to filter for relevance.
03.
Recommendation Engine
The LLM generates a personalized rationale for each recommendation, explaining *why* it fits the user's current mood.