RefineSnap — AI Image Generator

Turn a text prompt into a stunning HD image in seconds.

A full-stack AI image generation platform built with a vanilla JavaScript frontend and a Spring Boot backend powered by Spring AI. Type a prompt, hit generate, get a 1792×1024 HD image — no fuss.

---

Demo

---

Features

• HD image generation via OpenAI DALL·E 3 (1792×1024, natural style)
• Clean, responsive UI — no framework dependencies
• Spring AI ImageModel abstraction for easy provider swapping
• Fully configurable quality, dimensions, and style via OpenAiImageOptions
• RESTful API backend ready to plug into any frontend or mobile app

---

Tech Stack

Layer	Technology
Frontend	HTML5, CSS3, Vanilla JavaScript
Backend	Java 21, Spring Boot 3, Spring AI
AI Provider	OpenAI DALL·E 3
Build	Maven

---

Getting Started

Prerequisites

• Java 21+
• Maven
• OpenAI API key

Clone & run

git clone https://github.com/your-username/refine-snap-image-generator.git
cd refine-snap-image-generator

Set your API key

In src/main/resources/application.properties:

spring.ai.openai.api-key=YOUR_OPENAI_API_KEY

Start the server

mvn spring-boot:run

Open index.html in your browser or serve it with any static file server.

---

API Reference

POST /api/images/generate
Content-Type: application/json

{
  "prompt": "A futuristic city skyline at golden hour, photorealistic"
}

Response:

{
  "imageUrl": "https://..."
}

---

Architecture

The backend uses Spring AI's ImageModel abstraction to decouple the application from OpenAI's SDK directly. Swapping to another image provider (Stability AI, etc.) requires changing a single bean — no business logic changes.

Frontend (HTML/JS)
    └── POST /api/images/generate
          └── ImageService
                └── Spring AI ImageModel
                      └── OpenAI DALL·E 3

---

Contributing

PRs are welcome. Please open an issue first for major changes.

---

License

MIT

---

Built by [Your Name] · Open to full-stack and AI engineering roles · LinkedIn