matinsn2000's picture
Updated README
e12473b
---
title: Cloudzy AI Photo Management
emoji: ๐Ÿ“ธ
colorFrom: blue
colorTo: purple
sdk: docker
pinned: true
---
Cloudzy AI Photo Management
Cloudzy is a cloud-based photo management service with AI-powered analysis and semantic search. It allows users to upload photos, automatically analyzes them using AI models, and provides intelligent features such as semantic search and creative AI-generated insights.
๐Ÿš€ Objective
Build a photo management service with the following requirements:
Photo Upload and Storage
POST /upload: Receive a photo, store it, and generate a unique ID.
GET /photo/:id: Return metadata including upload date, file name, and AI-generated tags and captions.
AI-Powered Analysis
Each uploaded photo must use a vision model (local or API-based) to produce:
Tags (at least 5 relevant)
Caption (short descriptive sentence)
Embedding vector (for semantic search)
Semantic Search
GET /search?q=... : Return photos ranked by semantic similarity using embeddings.
Mandatory AI-Powered Smart Feature
Include at least one creative AI-powered feature, e.g.:
Album generation
Daily summaries
Emotion/facial recognition
AI Usage Report
Document where and how AI was used
Provide prompts and model inputs
Explain how model outputs were refined
๐Ÿ“‚ Example Workflow
Upload a photo via /upload.
The system generates:
Tags and captions using AI
Embeddings for semantic search
Retrieve photo metadata via /photo/:id.
Search for photos with /search?q=<query> to find semantically similar images.
Enjoy AI-powered smart features like automatic album generation or summaries.
๐Ÿค– AI Usage
This project integrates AI in multiple stages:
Tagging and Captioning: Vision model automatically labels photos and generates descriptive captions.
Embedding Generation: Produces vector representations for semantic search.
Creative AI Features: Enhances user experience, e.g., grouping similar photos into albums or summarizing daily uploads.
๐Ÿ“Œ Notes
The service is designed for mandatory AI usageโ€”all main features leverage AI models.
This README uses Hugging Face Spaces front matter to define metadata, including color, emoji, and SDK type.