Go Identiface

A simple library for identifying face with golang

Requirements

This library is based on go-face but with support for ID data type customization. So, this library require go-face to be used and go-face require dlib >= 19.10 and libjpeg development packages installed to run.

Quoting from here for go-face requirement is below.

Ubuntu 18.10+, Debian sid

Latest versions of Ubuntu and Debian provide suitable dlib package so just run:

# Ubuntu
sudo apt-get install libdlib-dev libblas-dev libatlas-base-dev liblapack-dev libjpeg-turbo8-dev --no-install-recommends
# Debian
sudo apt-get install libdlib-dev libblas-dev libatlas-base-dev liblapack-dev libjpeg62-turbo-dev --no-install-recommends

macOS

Make sure you have Homebrew installed.

brew install dlib

Windows

Make sure you have MSYS2 installed.

1. Run MSYS2 MSYS shell from Start menu

2. Run pacman -Syu and if it asks you to close the shell do that

3. Run pacman -Syu again

4. Run pacman -S mingw-w64-x86_64-gcc mingw-w64-x86_64-dlib

5. Set env variables

   • If you already have Go and Git installed and available in PATH uncomment set MSYS2_PATH_TYPE=inherit line in msys2_shell.cmd located in MSYS2 installation folder

   • Otherwise run pacman -S mingw-w64-x86_64-go git

6. Run MSYS2 MinGW 64-bit shell from Start menu to compile and use go-face

Other systems

Try to install dlib/libjpeg with package manager of your distribution or compile from sources. Note that go-face won't work with old packages of dlib such as libdlib18. Alternatively create issue with the name of your system and someone might help you with the installation process.

Docker

If you use Docker. I have created a sample Dockerfile for you here and you can customize according you need.

---

NOTE: Additional Information for Each Platform

go-face used C and C++ language for some case. So, you need to install C and C++ compiler to use this project as dependency (just for compile not for run), if you have no C and C++ compiler you will get error "face.Face is undefined" if you want to compile your golang project. Makesure you have C and C++ compiler installed but if you don't have any C and C++ compiler. Just install it with following command below:

• Ubuntu 18.10+, Debian sid

  sudo apt-get install g++ gcc --no-install-recommends

• Windows

  For windows follow this instructions

• MacOS

  brew install gcc

Usage

To use go-identiface to your Go code, use it as dependency below:

go get github.com/irdaislakhuafa/go-identiface@latest

Then import it on your Go project:

import "github.com/irdaislakhuafa/go-identiface"

Example

NOTE:

• Make sure you have installed all dependencies above and downloaded the sample assets with command make get-assets. This command required Makefile package.

package main

import (
	"fmt"
	"os"
	"path/filepath"

	"github.com/google/uuid"
	"github.com/irdaislakhuafa/go-identiface"
)

type user struct {
	id        string `json:"id,omitempty"`
	name      string `json:"name,omitempty"`
	imagePath string `json:"image_name,omitempty"`
}

var (
	assetsDir = "./assets"
	modelsDir = filepath.Join(assetsDir, "models")
	imagesDir = filepath.Join(assetsDir, "images")
)

// Run this code at root dir of project with `go run ./example/main.go`
func main() {
	// Imagine this is data users from database
	users := []user{
		{id: uuid.NewString(), name: "Tony Stark", imagePath: filepath.Join(imagesDir, "tony-stark.jpeg")},
		{id: uuid.NewString(), name: "Tzuyu", imagePath: filepath.Join(imagesDir, "tzuyu.jpg")},
	}

	getUserByID := map[string]user{}
	for _, user := range users {
		getUserByID[user.id] = user
	}

	// Initialize `identiface` [string] is data type of key for each image
	iFace, err := identiface.Init[string](modelsDir)
	if err != nil {
		panic(err)
	}
	defer iFace.Close()

	// Adding image datasets to `go-identiface`
	for _, user := range users {
		fileBytes, err := os.ReadFile(user.imagePath)
		if err != nil {
			panic(err)
		}

		if err := iFace.AddSingleDatasetFromBytes(user.id, fileBytes); err != nil {
			panic(err)
		}
	}

	// After adding datasets, don't forget to load them so they can be used by `go-identiface`. After this you can identify user by human face of image
	iFace.LoadDatasets()

	// Imagine this is an image file sent from client (ex. Mobile/Frontend/Etc)
	clientImg := "tony-stark3.jpeg"
	clienImgBytes, err := os.ReadFile(filepath.Join(imagesDir, clientImg))
	if err != nil {
		panic(err)
	}

	// After get image file from client then you can identify client by face of the image
	data, err := iFace.ClassifySingleFromBytes(clienImgBytes)
	if err != nil { // Will return error if face is not recognized or image contain multiple face
		panic(err)
	}

	// Imagine this is method to get detail data user from db
	user := getUserByID[data.ID]

	// Now you can identify user by face
	fmt.Printf("user: %#v\n", user)
}

You can see detail example usage of code here

Testing

To run testing code at identiface_test.go if you have installed all dependencies above depend of your Operating System you can run go test -v .../.. directly in root folder. But if you don't want to pollute your system with temporary dependencies to test code use the command below or you can modify test.Dockerfile according you need.

docker compose up test

Todo

• Handle identify for grey image