Table of Contents

• Environment Setup and Requirements Installation

  • Step 1: Set Up a Python Virtual Environment
  • Step 2: Install CUDA-Compatible PyTorch
  • Step 3: Install Additional Requirements
  • Summary of Commands

• Accessing Projects

  • 1st Assignment
  • 2nd Assignment
  • 3rd Assignment

Environment Setup and Requirements Installation

Here's a step-by-step guide to set up the environment and install requirements for your assignments, ensuring each step is covered from environment setup to dependencies installation.

Recommended IDE to run -

• VS code

Step 1: Set Up a Python Virtual Environment

A virtual environment helps isolate project-specific dependencies. Here's how to create one:

1. Navigate to your project directory or create a new directory for the assignments. clone this repository by the command

git clone https://github.com/yashbudhia/emmetra.git

2. Create the virtual environment using Python (replace env with your preferred name for the environment):

   python -m venv env

3. Activate the virtual environment:

   • On Windows:

     .\env\Scripts\activate

   • On macOS/Linux:

     source env/bin/activate

Step 2: Install CUDA-Compatible PyTorch

Since your assignments require PyTorch with CUDA support, install PyTorch for CUDA 11.8 specifically (replace 11.8 if you have a different CUDA version installed).

• . Use the following command to install PyTorch for CUDA 11.8:

  pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118

Step 3: Install Additional Requirements

Your assignments require additional packages beyond PyTorch, such as OpenCV, NumPy, ReportLab, and potentially TensorFlow for model-based denoising.

• . Install the dependencies from the requirements.txt file:

  pip install -r requirements.txt

This should display the installed packages, confirming that PyTorch, OpenCV, ReportLab, NumPy, and TensorFlow are available.

Summary of Commands

Here's a consolidated list for quick reference:

# 1. Set up and activate the virtual environment
python -m venv env
source env/bin/activate # On macOS/Linux
.\env\Scripts\activate # On Windows

# 2. Install PyTorch for CUDA 11.8 (or install specific version of pytorch for your version of cuda)
pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118

# 3. Install other requirements
pip install -r requirements.txt

# 4. Verify installed packages
pip list

Accessing Projects

1st Assignment

1. Go inside First Assignment Folder

cd 1st-assignment

2. Run Flask server by typing the command in the terminal

python app.py

3. The UI for Custom Image Signal Processing pipeline will appear on the web browser in the url localhost:5000

4. If you want to see OpenCV implementation - Go inside the folder

cd openCV-version

• And run flask server by typing the command in the terminal

python app.py

5. If you want to see Custom ISP code without any UI . Just Naviagate to Custom-ISP folder and run python ISP.py in the terminal

cd ..
cd Custom-ISP

• Feel free to tweak the parameters

2nd Assignment

For Running 2nd assignment - follow the steps

Step 1 : Navigate to 2nd-assignment folder if the environment is setup

cd emmetra/2nd-assignment

Step 2 : First go to ffdnet model folder and generate a denoised image

cd ffdnet-pytorch

• after reaching here run the command in terminal to denoise sample image obtained from the ISP from assignment 1(image.png)

python test_ffdnet_ipol.py --input image.png --noise_sigma 25 --add_noise True

• After the model has generated the image now go back

cd ..

• Now run

python main.py

• It will then generate reports of the image in various denoising and edge effects.

• You can configure the input by replacing image.png in inputs folder with your image (rename it to image.png) and in ffdnet folder too

• Remember - Run the FFDnetModel before running the main.py script.

Custom model usage -

• Custom model was still in development(unfortunately) so would recommend just leave it as is. anyway run

python main_processing.py

to generate reports using custom model

3rd Assignment

Overview Briefly explain the purpose of the assignment, its main functionality, and any core concepts or algorithms it implements.

Environment Setup and Requirements Installation

Step 1: Navigate to 3rd-assignment folder if the environment is setup

cd emmetra/3rd-assignment

Step 2: Running the Assignment

Include any instructions for running scripts, such as:

Run the main script:

python app.py

• The UI Will then appear on the Web Browser at localhost:5000

Step 3: Upload the images

• Upload the images at given exposures and obtain the HDR image. Sample inputs are present at inputs folder at 3rd-assignment folder.