Cartoonizer is an interactive web application that transforms regular images into cartoon-like visuals. Utilizing state-of-the-art models like CartoonGAN and White-box Cartoonization, the application ensures high-quality cartoon effects that are visually appealing and closely resemble hand-drawn art.
cartoon/
│
├── app.py # Main application file running the web server and handling requests
├── final.yaml # YAML configuration file
│
├── cartoonGAN/ # Contains implementation and assets related to the CartoonGAN model
│ ├── generator_latest.pth # Pre-trained generator model
│ └── transform.py # Transformation script using CartoonGAN
│
├── white_box_cartoonization/ # Contains implementation and assets for the White-box Cartoonization model
│ ├── cartoonize.py # Script to cartoonize images using White-box Cartoonization
│ └── saved_models/ # Directory containing model files for white box cartoonization
│
├── static/ # Directory to store static assets like images
│ ├── uploads/ # Sample uploaded files
│ └── cartoons/ # Sample cartoon images
│
└── templates/ # Contains HTML templates defining the structure and look of web pages
├── index.html # Main HTML template for the web application
- app.py: Main application file that runs the web server and handles user requests.
- cartoonGAN: Implementation and assets related to the CartoonGAN model.
- white_box_cartoonization: Implementation and assets for the White-box Cartoonization model.
- static: Stores static assets like images.
- templates: Contains HTML templates which define the structure and look of the web pages.
- final.yaml: A YAML configuration file, possibly related to the environment or other settings.
- Ensure that you have Anaconda or Miniconda installed.
- Navigate to the project directory.
- Create a new conda environment using the provided final.yaml file:conda env create -f final.yaml
- Activate the newly created environment:conda activate <env_name>
Replace <env_name>
with the environment name specified in the final.yaml file.
- Ensure you're in the project directory.
- Run the app.py script:python app.py
- Open a web browser and navigate to the address shown in the console (usually http://127.0.0.1:5000/).
- CartoonGAN: Employs Generative Adversarial Networks, trained on a vast dataset of real-world images and their cartoon versions.
- White-box Cartoonization: A model providing a unique cartoon style, allowing users to choose between different cartoonization effects.
- Nitya Parikh - UTA ID: 1002053645, NET ID: nxp3645
- Mayank Vekariya - UTA ID: 1002078999, NET ID: mxv8999