CropSense - Face Detection in C++

(C++ port of CropSense-Face-Detection)

CropSense is a tool to crop images based on various bounding box sizes, such as upper body, face, or full body. It utilizes a pre-trained SSD face detection model to accurately identify faces within images and performs precise cropping into a 1:1 ratio based on the detected regions. Useful for creating datasets for model training.

Requirements

• CMake 3.10+
• OpenCV
• Ninja (optional)

How to Build

Windows

1. Open Native Tools Command Prompt for VS 2022
2. Navigate to your project folder:

   cd your_project_path

3. Clear existing build folder (if any):

   rmdir /s /q build

4. Configure and build with CMake:

   cmake -B build -DOpenCV_DIR="C:/your_path/opencv/build"
   cmake --build build

5. Or with Ninja (faster):

   cmake -B build -G Ninja -DOpenCV_DIR="C:/your_path/opencv/build"
   cmake --build build

Linux (Ubuntu/Debian)

1. Install OpenCV:

   sudo apt-get install -y libopencv-dev ninja-build

2. Configure and build:

   cmake -B build -G Ninja
   cmake --build build

Usage

1. Place your images in the input folder
2. Copy config.json, deploy.prototxt.txt, and res10_300x300_ssd_iter_140000.caffemodel next to the executable
3. Run main.exe (Windows) or ./main (Linux)
4. Select crop type and preview option

License

This project is licensed under the MIT License.