Image Colorization

Description

This project aims to implement an image colorization algorithm using deep learning techniques.

Features

• Convert grayscale images to colorized images
• Support for various image formats (e.g., JPEG, PNG)
• User-friendly interface using Streamlit, a simple web app framework for Python

Installation

1. Clone the repository:

   git clone https://github.com/NithinHD/Image-colorization.git

2. Install the required dependencies:

   pip install -r requirements.txt

Usage

1. Navigate to the project directory:

   cd Image-Colorization

2. Run the colorization app:

   streamlit run app.py

   Open the provided URL in a web browser and follow the instructions to colorize images.

3. Run the notebook:

   jupyter notebook

   Open the Image Colorization.ipynb notebook and follow the instructions to colorize images.

Description of the Model

• Input Layer:
  • This layer serves as the entry point for the input data, typically images or feature maps.
• Convolutional Layer 1 (Conv1):
  • Convolution operation extracts features from the input using a set of learnable filters. ReLU activation function introduces non-linearity to the model.
  • Batch normalization normalizes the activations of the previous layer, which helps in training stability and faster convergence.
• Convolutional Layer 2 (Conv2):
  • Similar to Conv1, this layer extracts higher-level features from the input.
  • ReLU activation and batch normalization are applied for non-linearity and normalization.
• Convolutional Layer 3 (Conv3):
  • Another convolutional layer to further extract complex features.
  • ReLU activation and batch normalization are applied.
• Convolutional Layer 4 (Conv4):
  • Continues the feature extraction process.
  • ReLU activation and batch normalization.
• Convolutional Layer 5 (Conv5):
  • Further extracts features with deeper representation.
  • ReLU activation and batch normalization.
• Convolutional Layer 6 (Conv6):
  • Continues the feature extraction process. ReLU activation and batch normalization.
• Convolutional Layer 7 (Conv7):
  • Extracts more abstract features.
  • ReLU activation and batch normalization.
• Deconvolutional Layer (Conv8):
  • Deconvolution (also known as transposed convolution) is used for upsampling or generating feature maps with larger spatial resolutions.
  • ReLU activation is applied.
• Softmax Layer:
  • Softmax activation function is typically used for multi-class classification to compute probabilities for each class.
• Convolutional Layer for Decoding (class8_ab):
  • This layer is used for decoding or generating output.

Results

Contributing

Contributions are welcome! If you find any issues or have suggestions for improvement, please open an issue or submit a pull request.

License

This project is licensed under the MIT License. See the LICENSE file for more information.