Kenan Paralija
Published © CERN-OHL

Simple Face search Function with ESP32Cam

ESP32Cam and 2Servos to search and following the Face- The main objective of this project is to perform detection and tracking of faces.

IntermediateProtip3 hours228
Simple Face search Function with ESP32Cam

Things used in this project

Hardware components

SG90 Micro-servo motor
SG90 Micro-servo motor
ESP32 Camera Module Development Board
M5Stack ESP32 Camera Module Development Board
ESP32Cam controls the servo

Hand tools and fabrication machines

Multitool, Screwdriver
Multitool, Screwdriver


Read more


Circuit diagram

In a design where ESP32Cam is supposed to control the servo directly, we get other problems like interference and connections with the servo.



The main objective of this program is to perform detection and tracking of faces from the real-time input video. The servos control panning and tilting the webcam mounted on it. The webcam’s position changes according to the movement of the object or person.
//ESP32 camera: face detection
//NO body movement, Servo on ESP32
//how to detect faces using the ESP32cam
//It can detect a human face with the model MTMN.
//The output is the human face coordinates if there is one in the image.
//left top, right down, landmarks

//ESP32 Camera Driver, Camera OV2640/1600 x 1200/Len Size:1/4"

#include "esp_camera.h"
#include "fd_forward.h"
#include "fb_gfx.h"
#include "soc/soc.h"             // disable brownout problems
#include "soc/rtc_cntl_reg.h"    // disable brownout problems
#include <ESP32Servo.h>

//***** Servo def
#define DUMMY_SERVO1_PIN 12     //We need to create 2 dummy servos.
#define DUMMY_SERVO2_PIN 13     //So that ESP32Servo library does not interfere with pwm channel and timer used by esp32 camera.
#define PAN_PIN 14
#define TILT_PIN 15

Servo dummyServo1;
Servo dummyServo2;
Servo panServo;
Servo tiltServo;
// Published values for SG90 servos; adjust if needed
int minUs = 500;
int maxUs = 2400;
//**** Servo the end

int  posH;  //Position Panoram
int  posV;  //Position Tilt
int altPosH;
bool toNul; //servo moves to Null

//#define LED_BUILTIN 4 // No #define LED_PIN
#define PWDN_GPIO_NUM     32
#define RESET_GPIO_NUM    -1
#define XCLK_GPIO_NUM      0
#define SIOD_GPIO_NUM     26
#define SIOC_GPIO_NUM     27
#define Y9_GPIO_NUM       35
#define Y8_GPIO_NUM       34
#define Y7_GPIO_NUM       39
#define Y6_GPIO_NUM       36
#define Y5_GPIO_NUM       21
#define Y4_GPIO_NUM       19
#define Y3_GPIO_NUM       18
#define Y2_GPIO_NUM        5
#define VSYNC_GPIO_NUM    25
#define HREF_GPIO_NUM     23
#define PCLK_GPIO_NUM     22

//++++++++++++++++++++++++++++++++++++++++++++ function called draw_face_boxes that is used to display a box around a detected face
static void draw_face_boxes(dl_matrix3du_t *image_matrix, box_array_t *boxes)
  int x, y, w, h, i, half_width, half_height;  
  fb_data_t fb;
  fb.width = image_matrix->w;
  fb.height = image_matrix->h; = image_matrix->item;
  fb.bytes_per_pixel = 3;
  fb.format = FB_BGR888;
  for (i = 0; i < boxes->len; i++) {

    // finding face centre...
    x = ((int)boxes->box[i].box_p[0]);
    w = (int)boxes->box[i].box_p[2] - x + 1;
    half_width = w / 2;
    //++++++++++++++++++++++++++ Center PAN is 160 +++++++++++++++++++++++++++++++++++++++++++++++
    int face_center_pan = x + half_width; // image frame face centre x co-ordinate

    // definition of servo move
    y = (int)boxes->box[i].box_p[1];
    h = (int)boxes->box[i].box_p[3] - y + 1;
    half_height = h / 2;
    posH =posH + (160 - face_center_pan)/7; //was 6,
     if ((altPosH-posH)>0) 
     toNul=false;  //// servo move to 0 grad
        toNul=true; // servo move to plus 180
    if (posH>170) 
    if (posH<20) 
    Serial.printf("Center detected at %d dots\n", face_center_pan);
  Serial.printf("H%d \n", posH);
    //++++++++++++++++++++++++ Center TILT is 120 ++++++++++++++++++++++++++++++++++++++++++++++++++   
    int face_center_tilt = y + half_height;  // image frame face centre y co-ordinate
    posV =posV - (120 - face_center_tilt)/7;  //
    if (posV>130) 
      posV=130;   //LIMIT UP;
    if (posV<50) 
      posV=50;    //LIMIT DOWN
     Serial.printf("V%d \n", posV);

mtmn_config_t mtmn_config = {0};
 // configurations of MTMN that will be used to detect the faces -
 //using the default configurations
int noDetection = 0;
  //define a variable that will count how many times we have not detected faces. 
  //We will initialize it with the value zero and increment it every time faces are not detected in a frame.

void setup() {
   if (!initCamera()) {
    Serial.printf("Failed to initialize camera...");

  //Serial2.begin(9600, SERIAL_8N1, RXD2, TXD2);
  // Serial for ARDUINO auf  Pins Rx>13, TX>15
  //  delay(1000);
     //  if(Serial2){
     //  Serial.println("******Serial2 successfully set up******");   } 
   //pinMode (LED_BUILTIN, OUTPUT); // initialize digital pin 4 LED_BUILTIN as an output.
  mtmn_config = mtmn_init_config();
  //****** Servo
  panServo.setPeriodHertz(50);      // Standard 50hz servo
  tiltServo.setPeriodHertz(50);      // Standard 50hz servo
  panServo.attach(PAN_PIN, minUs, maxUs);
  tiltServo.attach(TILT_PIN, minUs, maxUs);
  panServo.write(posH); //pan angle 90 start
  tiltServo.write(posV);  //tilt angle 90 start
  //******** The End

void loop() {
  camera_fb_t * frame;
  frame = esp_camera_fb_get();    
  //call the esp_camera_fb_get function to get an image from the camera

  dl_matrix3du_t *image_matrix = dl_matrix3du_alloc(1, frame->width, frame->height, 3);
    //As output, this function will return a pointer to the allocated matrix struct.
    //to convert the captured image to RGB888 format . Sign "->" allows to access elements in Structures
  fmt2rgb888(frame->buf, frame->len, frame->format, image_matrix->item);
  //This function call will allow the image buffer to be reused again, 
  box_array_t *boxes = face_detect(image_matrix, &mtmn_config);
  //A box_array_t type value contains face boxes -coordinates: left top, right down, landmarks 
  if (boxes != NULL) {
    noDetection = 0;
    Serial.printf("Faces detected at %d \n", millis());
    draw_face_boxes(image_matrix, boxes);   //print command for move or pan camera
     noDetection = noDetection+1;
     Serial.printf("Faces not not detected %d times \n", noDetection);
    //  As a timer for the search function, we use a variable "noDetection", 
        //which accumulates with each unsuccessful attempt at face detection.
    case 10:  Serial.printf("Case10 Nodetected at %d \n", millis()); break; // execute the corresponding function wenn after 5 times NOdetektion
    case 40: {
              Serial.printf("Case40 Nodetected at %d \n", millis());   //xxxxxxxxx
              noDetection = 0; // stop
    default:  break;
     if (toNul) 
     //block of code to be executed if the condition is true
      //  posH++; speed search
      if (posH>170) 
    // //  posH--; speed search
    if (posH<10) 

bool initCamera() {
  camera_config_t config;
  config.ledc_channel = LEDC_CHANNEL_0;
  config.ledc_timer = LEDC_TIMER_0;
  config.pin_d0 = Y2_GPIO_NUM;
  config.pin_d1 = Y3_GPIO_NUM;
  config.pin_d2 = Y4_GPIO_NUM;
  config.pin_d3 = Y5_GPIO_NUM;
  config.pin_d4 = Y6_GPIO_NUM;
  config.pin_d5 = Y7_GPIO_NUM;
  config.pin_d6 = Y8_GPIO_NUM;
  config.pin_d7 = Y9_GPIO_NUM;
  config.pin_xclk = XCLK_GPIO_NUM;
  config.pin_pclk = PCLK_GPIO_NUM;
  config.pin_vsync = VSYNC_GPIO_NUM;
  config.pin_href = HREF_GPIO_NUM;
  config.pin_sscb_sda = SIOD_GPIO_NUM;
  config.pin_sscb_scl = SIOC_GPIO_NUM;
  config.pin_pwdn = PWDN_GPIO_NUM;
  config.pin_reset = RESET_GPIO_NUM;
  config.xclk_freq_hz = 20000000;
  config.pixel_format = PIXFORMAT_JPEG;
  config.frame_size = FRAMESIZE_QVGA;
  config.jpeg_quality = 10;
  config.fb_count = 1;
  esp_err_t result = esp_camera_init(&config);
  if (result != ESP_OK) {
    return false;
  return true;


Kenan Paralija

Kenan Paralija

8 projects • 4 followers
I started my career by designing analog and digital circuits, moved to telecommunication and finally as a IT/TK lecturer.