GoFace

A university project in Golang for face recognition.

Running GoFace

Using Go Command

To run the GoFace project with Go installed, open a command prompt or terminal and navigate to the main project directory. Then execute the following command:

go run ./cmd/GoFace

This method requires Go to be installed on your system.

Using the Executable

If you prefer not to install Go or are working on a system without Go, you can use the GoFace.exe executable for ease. Simply replace the command go run ./cmd/GoFace with `./GoFace.exe`` in all usage instructions. For instance:

./GoFace.exe <arguments>

This executable allows you to run the GoFace project directly, without needing Go installed.

CLI Usage

The GoFace CLI provides a simple interface to run face detection on images. You can specify an individual image or process all images within a directory. Additionally, you have the option to draw either circles or rectangles around detected faces, specify the output directory, get help information, and measure performance.

Commands

1. Detect Face in a Single Image: To detect faces in a single image, provide the image file name (relative to the images directory) as an argument.

   go run ./cmd/GoFace -o outputdir <image_file_name>

   For example, if you have an image named ja.jpg in the images directory:

   go run ./cmd/GoFace -o outputdir ja.jpg

2. Detect Faces in All Images in a Directory:

   To run face detection on all images in the images directory, use the all argument.

   go run ./cmd/GoFace -o outputdir all

3. Option to Draw Circles or Rectangles:

   By default, rectangles are drawn around detected faces. You can choose to draw circles instead by using the -c flag.

   • For a single image:

     go run ./cmd/GoFace -c -o outputdir <image_file_name>

   • For all images:

     go run ./cmd/GoFace -c -o outputdir all

4. Display Help Information:

   To display help information about the available commands and flags, use the -h flag.

   go run ./cmd/GoFace -h

5. Measure Performance:

   To measure and display the performance of the face detection process, use the -p flag.

   • For a single image:

     go run ./cmd/GoFace -p -o outputdir <image_file_name>

   • For all images:

     go run ./cmd/GoFace -p -o outputdir all

Output

By default the processed images will be saved in the output directory with the same file names as the input images.

Running Tests

To execute the tests for this project, use the following command in the terminal or command prompt:

go test -v ./tests    

his command will run all the tests located in the tests directory. The -v flag provides verbose output, showing detailed information about each test case.

If you prefer a more concise output, indicating only whether the tests passed or failed, you can omit the -v flag:

go test ./tests

Acknowledgments and References

Pigo - Pure Go Face Detection Library

This project utilizes Pigo, a pure Go library for face detection, as a key component in the face detection process. Pigo provides an efficient and easy-to-use way to detect faces in images without any external dependencies.

For more information about Pigo and its usage, visit the Pigo GitHub repository.

Sources for Test Images

For testing the face detection capabilities, you can obtain images from various sources. Here are some suggestions:

• Dorian Studio - Specializes in school photography, offering a range of portraits and group photos.
• Adobe Stock - A rich resource for high-quality stock images, including a wide range of facial photos.
• Miedzyrzecz Photography - An informational portal featuring news and information about Międzyrzecz and the surrounding region.

Note: Ensure to respect copyright and usage rights when using images from these sources.