[Ubuntu] [Python] Facial organ detection using dlib

Facial organ detection that is currently in dlib but not in OpenCV.

For the time being, I forcibly colored it, but there must be a smarter way. Or rather, I think it's just that the reference isn't loaded properly. .. ..

The video is below. It's embarrassing to show my face, so I use face detection to hide it. https://www.youtube.com/watch?v=s2YtXqcBuPY [](https://www.youtube.com/watch? v = s2YtXqcBuPY)

The source code is below. To make it work, put the trained data of facial organs in the same directory as the py file. Must be placed. →shape_predictor_68_face_landmarks.dat（http://dlib.net/files/shape_predictor_68_face_landmarks.dat.bz2）

#!/usr/bin/env python
# -*- coding: utf-8 -*-

'''
face_landmark_detector.py.

Usage:
  face_landmark_detector.py [<video source>] [<resize rate>] [<privacy mask>]
'''

import sys
import dlib
import cv2
import time
import copy

try:
    fn = sys.argv[1]
    if fn.isdigit() == True:
        fn = int(fn)
except:
    fn = 0

try:
    resize_rate = sys.argv[2]
    resize_rate = int(resize_rate)
except:
    resize_rate = 1

try:
    privacy_mask = sys.argv[3]
    privacy_mask = int(privacy_mask)
except:
    privacy_mask = 0

predictor_path = "./shape_predictor_68_face_landmarks.dat"

detector = dlib.get_frontal_face_detector()
predictor = dlib.shape_predictor(predictor_path)

video_input = cv2.VideoCapture(fn)

while(video_input.isOpened() == True):
    ret, frame = video_input.read()
    temp_frame = copy.deepcopy(frame)

    #Target frame reduction to reduce processing load (when argument is specified)
    height, width = frame.shape[:2]
    temp_frame = cv2.resize(frame, (int(width/resize_rate), int(height/resize_rate)))

    #Face detection
    start = time.time()
    dets = detector(temp_frame, 1)
    elapsed_time = time.time() - start
    print ("detector processing time:{0}".format(elapsed_time)) + "[sec]"

    for k, d in enumerate(dets):
        print("Detection {}: Left: {} Top: {} Right: {} Bottom: {}".format(
            k, d.left(), d.top(), d.right(), d.bottom()))

        #Facial organ detection
        start = time.time()
        shape = predictor(temp_frame, d)
        elapsed_time = time.time() - start
        print ("predictor processing time:{0}".format(elapsed_time)) + "[sec]"

        #drawing
        rect_offset = 20
        if privacy_mask == 1:
            cv2.rectangle(frame, (int(d.left() * resize_rate) - rect_offset, int(d.top() * resize_rate) - rect_offset), \
                (int(d.right() * resize_rate) + rect_offset, int(d.bottom() * resize_rate) + rect_offset), (255, 255, 255), -1)

        for shape_point_count in range(shape.num_parts):
            shape_point = shape.part(shape_point_count)
            if shape_point_count < 17: # [0-16]:Contour
                cv2.circle(frame, (int(shape_point.x * resize_rate), int(shape_point.y * resize_rate)), 2, (0, 0, 255), -1)
            elif shape_point_count < 22: # [17-21]Eyebrows (right)
                cv2.circle(frame, (int(shape_point.x * resize_rate), int(shape_point.y * resize_rate)), 2, (0, 255, 0), -1)
            elif shape_point_count < 27: # [22-26]Eyebrows (left)
                cv2.circle(frame, (int(shape_point.x * resize_rate), int(shape_point.y * resize_rate)), 2, (255, 0, 0), -1)
            elif shape_point_count < 31: # [27-30]Dorsal nasal
                cv2.circle(frame, (int(shape_point.x * resize_rate), int(shape_point.y * resize_rate)), 2, (0, 255, 255), -1)
            elif shape_point_count < 36: # [31-35]Ala of nose, tip of nose
                cv2.circle(frame, (int(shape_point.x * resize_rate), int(shape_point.y * resize_rate)), 2, (255, 255, 0), -1)
            elif shape_point_count < 42: # [36-4142 eyes 47)
                cv2.circle(frame, (int(shape_point.x * resize_rate), int(shape_point.y * resize_rate)), 2, (255, 0, 255), -1)
            elif shape_point_count < 48: # [42-47]Eyes (left)
                cv2.circle(frame, (int(shape_point.x * resize_rate), int(shape_point.y * resize_rate)), 2, (0, 0, 128), -1)
            elif shape_point_count < 55: # [48-54]Upper lip (upper contour)
                cv2.circle(frame, (int(shape_point.x * resize_rate), int(shape_point.y * resize_rate)), 2, (0, 128, 0), -1)
            elif shape_point_count < 60: # [54-59]Lower lip (lower contour)
                cv2.circle(frame, (int(shape_point.x * resize_rate), int(shape_point.y * resize_rate)), 2, (128, 0, 0), -1)
            elif shape_point_count < 65: # [60-64]Upper lip (lower contour)
                cv2.circle(frame, (int(shape_point.x * resize_rate), int(shape_point.y * resize_rate)), 2, (0, 128, 255), -1)
            elif shape_point_count < 68: # [65-67]Lower lip (upper contour)
                cv2.circle(frame, (int(shape_point.x * resize_rate), int(shape_point.y * resize_rate)), 2, (128, 255, 0), -1)

    cv2.imshow('face landmark detector', frame)

    c = cv2.waitKey(50) & 0xFF

    if c==27: # ESC
        break

video_input.release()
cv2.destroyAllWindows()

that's all.