【最快人臉檢測模型開源】libfacedetection開源

來自於仕琪老師的人臉檢測庫，可以實現高達1500fps的人臉檢測
算法源碼已經開源

1.簡介

libfacedetection是一個優秀的人臉檢測算法庫，在pc和樹莓派上都能達到十分優秀的檢測結果：

windows上的表現

Method	Time	FPS	Time	FPS
	X64	X64	X64	X64
	Single-thread	Single-thread	Multi-thread	Multi-thread
OpenCV Haar+AdaBoost (640x480)	–	–	12.33ms	81.1
cnn (CPU, 640x480)	64.21ms	15.57	15.59ms	64.16
cnn (CPU, 320x240)	15.23ms	65.68	3.99ms	250.40
cnn (CPU, 160x120)	3.47ms	288.08	0.95ms	1052.20
cnn (CPU, 128x96)	2.35ms	425.95	0.64ms	1562.10

• OpenCV Haar+AdaBoost runs with minimal face size 48x48
• Face detection only, and no landmark detection included.
• Minimal face size ~12x12
• Intel® Core™ i7-7700 CPU @ 3.6GHz.

嵌入式設備Raspberry Pi 3 B+上的表現

Method	Time	FPS	Time	FPS
	Single-thread	Single-thread	Multi-thread	Multi-thread
cnn (CPU, 640x480)	512.04ms	1.95	174.89ms	5.72
cnn (CPU, 320x240)	123.47ms	8.10	42.13ms	23.74
cnn (CPU, 160x120)	27.42ms	36.47	9.75ms	102.58
cnn (CPU, 128x96)	17.78ms	56.24	6.12ms	163.50

• Face detection only, and no landmark detection included.
• Minimal face size ~12x12
• Raspberry Pi 3 B+, Broadcom BCM2837B0, Cortex-A53 (ARMv8) 64-bit SoC @ 1.4GHz

---

近日算法的源碼已經開源：

2.源碼

源碼中主要包含了四個文件，包括了模型數據、人臉檢測cnn模型的定義和卷積操作的實現等：

facedetectcnn-floatdata.cpp   //模型數據
facedetectcnn-int8data.cpp   //模型數據
facedetectcnn-model.cpp	     //cnn定義
facedetectcnn.cpp            //基本操作定義
facedetectcnn.h

---

1.使用樣例

API是這樣的：int * facedetect_cnn(unsigned char * result_buffer, //buffer memory for storing face detection results, !!its size must be 0x20000 Bytes!! unsigned char * rgb_image_data, int width, int height, int step); //input image, it must be RGB (three-channel) image!

//copy from https://github.com/ShiqiYu/libfacedetection/blob/master/example/libfacedetectcnn-example.cpp
#include <stdio.h>
#include <opencv2/opencv.hpp>
#include "facedetectcnn.h"

//定義緩沖區大學. Do not change the size!
#define DETECT_BUFFER_SIZE 0x20000
using namespace cv;

int main(int argc, char* argv[])
{
    if(argc != 2)
    {
    	//圖像的路徑輸入
        printf("Usage: %s <image_file_name>\n", argv[0]);
        return -1;
    }

	//load an image and convert it to gray (single-channel)
	Mat image = imread(argv[1]); 
	if(image.empty())
	{
		fprintf(stderr, "Can not load the image file %s.\n", argv[1]);
		return -1;
	}

	int * pResults = NULL; 
    //pBuffer is used in the detection functions.
    //If you call functions in multiple threads, please create one buffer for each thread!
    unsigned char * pBuffer = (unsigned char *)malloc(DETECT_BUFFER_SIZE);
    if(!pBuffer)
    {
        fprintf(stderr, "Can not alloc buffer.\n");
        return -1;
    }
	

	///////////////////////////////////////////
	// CNN face detection 
	// Best detection rate
	//////////////////////////////////////////
	//!!! The input image must be a RGB one (three-channel)
	//!!! DO NOT RELEASE pResults !!!
	pResults = facedetect_cnn(pBuffer, (unsigned char*)(image.ptr(0)), image.cols, image.rows, (int)image.step);
	//這個api用於檢測

    printf("%d faces detected.\n", (pResults ? *pResults : 0));
	Mat result_cnn = image.clone();
	//print the detection results
	for(int i = 0; i < (pResults ? *pResults : 0); i++)
	{
        
        //返回值position坐標點
        short * p = ((short*)(pResults+1))+142*i;
		int x = p[0];
		int y = p[1];
		int w = p[2];
		int h = p[3];
		//int neighbors = p[4];
		int confidence = p[4]; 
		int angle = p[5];

		//printf("face_rect=[%d, %d, %d, %d], neighbors=%d, angle=%d\n", x,y,w,h,neighbors, angle); 
		//update neifhbors to confidence,05302019
		printf("face_rect=[%d, %d, %d, %d], confidence=%d, angle=%d\n", x,y,w,h,confidence, angle);
		rectangle(result_cnn, Rect(x, y, w, h), Scalar(0, 255, 0), 2);   //畫框框
	}
	imshow("result_cnn", result_cnn);

	waitKey();

    //release the buffer
    free(pBuffer);

	return 0;
}

注：
1.可以使用avx2指令集編譯得到更大的加速。
2.於老師曾經寫了著名的《OpenCV入門教程》

ref:
https://blog.csdn.net/m0_37733057/article/details/68059552
https://blog.csdn.net/CV_Jason/article/details/78819088 https://blog.csdn.net/CV_Jason/article/details/78819088
https://blog.csdn.net/sinat_31425585/article/details/77891844

pic from pexels.com