https://github.com/ysskrishna/pixelmotion

A full-stack application built using React, TailwindCSS, FastAPI, Redis, Redis Queue, Docker, and WebSockets. It allows users to generate videos from images with a single click while providing real-time progress updates during the video generation process. The generated videos include animations and background music.
https://github.com/ysskrishna/pixelmotion

distributed-systems docker docker-compose fastapi package-creation pillow publisher-subscriber pubsub python python-package react reactjs redis redis-queue rq server-sent-events storage-management tailwindcss websocket

Last synced: 2 days ago
JSON representation

A full-stack application built using React, TailwindCSS, FastAPI, Redis, Redis Queue, Docker, and WebSockets. It allows users to generate videos from images with a single click while providing real-time progress updates during the video generation process. The generated videos include animations and background music.

• Host: GitHub
• URL: https://github.com/ysskrishna/pixelmotion
• Owner: ysskrishna
• License: agpl-3.0
• Created: 2025-01-21T10:43:10.000Z (6 days ago)
• Default Branch: main
• Last Pushed: 2025-01-24T14:10:22.000Z (3 days ago)
• Last Synced: 2025-01-24T14:24:14.133Z (3 days ago)
• Topics: distributed-systems, docker, docker-compose, fastapi, package-creation, pillow, publisher-subscriber, pubsub, python, python-package, react, reactjs, redis, redis-queue, rq, server-sent-events, storage-management, tailwindcss, websocket
• Language: Python
• Homepage:
• Size: 42.7 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# PixelMotion - Image to video generation tool

A full-stack application built using React, TailwindCSS, FastAPI, Redis, Redis Queue, Docker, and WebSockets. It allows users to generate videos from images with a single click while providing real-time progress updates during the video generation process. The generated videos include animations and background music.

Check out sample generated videos in the [media/samples](./media/samples) folder.

## Media

https://github.com/user-attachments/assets/2e3887e4-2f59-4cb3-97ad-1dd0d65cb20c

  

### Landing Page



### Pipeline Running



### Generated Video



### Download Video Feature



### Storage Dashboard



### Multiple Images Processing



## Architecture Diagram



## Features

- Generate videos from images with a single click

- Real-time progress updates during the video generation process

- Animations and background music added to the videos

- Clean and user-friendly interface

- Secure and scalable backend

- Containerized deployment

## Architecture Highlights

- **Microservices Architecture**: The application is built using multiple independent services:

  - Frontend Service (React)

  - Backend API Service (FastAPI)

  - AI Engine Worker Service

  - Redis Service for message queuing and caching

  - MinIO Service for object storage

  

- **Backend Architecture**:

  - FastAPI for high-performance async API

  - Event-driven communication using WebSockets and Redis Pub/Sub

  - Background task processing with Redis Queue

  - Task status tracking and progress updates

  - Secure file storage and retrieval

  - Real-time updates between services

- **Modular Code Organization**:

  - Shared code packaged as `pm_common` Python package

  - Common utilities for Redis, S3, and task management

  - Reusable components across services

- **Scalable Infrastructure**:

  - Docker containerization for all services

  - Redis Queue for distributed task processing

  - MinIO S3-compatible object storage

- **Modern Frontend**:

  - React with TailwindCSS for responsive UI

  - Real-time WebSocket integration

  - Clean component architecture

## Techstack used: 

- React

- TailwindCSS

- Nginx

- WebSockets

- FastAPI

- Redis

- Redis Queue

- Pillow

- MoviePy

- MinIO

- Docker

- Docker Compose

- Python

## Installation

1. Clone the repository

2. Run `docker compose -f docker-compose.yml up` to start the all instances

3. Run `docker compose -f docker-compose.yml down` to stop the all instances

4. Run `docker compose -f docker-compose.yml up --build` to build and start the all instances

5. Frontend will be available at `http://localhost:80/`

6. Backend will be available at `http://localhost:8000/`

7. Minio Object Storage will be available at `http://localhost:9001/`

   - Username: `minioadmin`

   - Password: `minioadminpassword`

8. Redis will be available at `http://localhost:6379/`

## License

Copyright (c) 2024 Y. Siva Sai Krishna

This project is licensed under the AGPL-3.0 License - see the LICENSE file for details.