A CLI tool to run computer vision algorithms.
Currently, only the homography algorithm is implemented.

• Run homography transformation on images using OpenCV.
• Specify source and destination points via command line or a coordinates file.
• Output the transformed image to a file or display it in a window.

1. Clone the repository:

   git clone https://github.com/prash-red/VisionAlgorithms.git
   cd VisionAlgorithms

2. Install dependencies:

   • CMake (>= 3.14)
   • OpenCV (>= 4.5)
   • CLI11 (>= 2.1)
   • CUDA (optional, for GPU support)

   OpenCV is recommended to built from source

3. Build the project:

   3.1 For CPU

   cmake -S ./ -B build
   cmake --build build   

   3.2 For GPU (requires CUDA)

   cmake -S ./ -B build -DENABLE_CUDA=ON
   cmake --build build

Run the homography algorithm:

./build/cv homography -f <image_path> -s <x0> <y0> <x1> <y1> <x2> <y2> <x3> <y3> -d <x0'> <y0'> <x1'> <y1'> <x2'> <y2'> <x3'> <y3'>

Or use a coordinates file:

./build/cv homography -f <image_path> -c <coords_file>

• -f, --file : Path to the input image (required)
• -s, --source : 4 source points (8 integers, separated by spaces)
• -d, --destination : 4 destination points (8 integers, separated by spaces)
• -c, --coords-file : File with 4 source and 4 destination coordinates (each line: x,y)
• -o, --output-file : Output image file (optional; if not set, displays the result)
• --cuda : Use CUDA for processing (optional; requires CUDA support)

Example: An example image with coordinates has been provided in the test folder

./build/cv homography -f tests/test_data/homography/document.jpg -c tests/test_data/homography/coords.txt