Skip to content

tejassinghbhati/bakashi_studio

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

17 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Bakashi Studio — Neural Style Transfer

Transforming your reality into a masterpiece through Neural Style Transfer.


Overview

Bakashi Studio is a high-performance web application for artistic exploration.
It uses Deep Learning and Neural Style Transfer to transform real-time webcam feeds and uploaded images into works of art inspired by masters like Van Gogh, Monet, and Picasso.

The interface follows a classical museum aesthetic, blending modern AI with timeless fine art.


Key Features

  • Real-time Artistic Vision — Apply 10+ styles to live webcam feed with low latency
  • Image Laboratory — Upload photos and reimagine them artistically
  • Precision Controls — Adjustable artistic intensity
  • Personal Museum — Capture, preview, and download creations
  • Premium UI — Ornate design with gold accents and Libre Caslon Text
  • High Performance — FastAPI + PyTorch optimized pipeline

Artistic Styles

Style Inspiration
Starry Night Vincent van Gogh
The Scream Edvard Munch
Great Wave Hokusai
Mosaic Ancient Roman Art
Monet Claude Monet
Udnie Francis Picabia

Tech Stack

Frontend

  • React 18 (TypeScript)
  • Tailwind CSS + Custom CSS
  • WebSockets
  • Vite

Backend

  • Python 3.10+
  • FastAPI
  • PyTorch & TorchVision
  • OpenCV & PIL

Installation & Setup

Prerequisites

  • Python 3.10+
  • Node.js 18+
  • Optional: NVIDIA GPU (CUDA)

Clone Repository

git clone https://github.com/tejassinghbhati/Bakashi-Studio.git
cd Bakashi-Studio

1. Run with One Click (Windows)

Double-click the start_app.bat file in the root directory.

2. Manual Setup

Backend:

cd backend
python -m venv venv
venv\Scripts\activate
pip install -r requirements.txt
python main.py

Frontend:

cd frontend
npm install
npm run dev

Deployment Configuration

To make the app work on the public internet (Vercel + Render), you must link them.

1. Backend (Render)

Deploy the backend folder to Render as a Web Service.

  • Build Command: sh render_build.sh
  • Start Command: uvicorn main:app --host 0.0.0.0 --port $PORT

2. Frontend (Vercel)

Deploy the frontend folder to Vercel. Crucial: You must add the following Environment Variables in your Vercel Project Settings so the frontend knows where the backend is.

Variable Name Value Example Note
VITE_API_URL https://your-app-name.onrender.com No trailing slash
VITE_WS_URL wss://your-app-name.onrender.com/ws/style Must be wss:// for secure SSL

After adding these, Redeploy your Vercel project.

About

Bakashi Studio is a premium, high-performance web application designed for artistic exploration. It leverages Deep Learning and Neural Style Transfer to transform real-time webcam feeds and uploaded images into stunning works of art, inspired by legendary masters like Van Gogh, Monet, and Picasso.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors