How do I align and crop the images located in subdirectories for face recognition?

Question

So, I have a got a folder with 132 subfolders inside it with name of the person as folder name. Each subfolders has 5 face images. I want to loop through all the subfolders and align and crop the image for face recognition and store all the subfolders on a new folder named 'aligned_face'. I have found a code for face cropping and alignment. My question is, how do I use this code to loop through my 132 subfolders and store all the aligned and cropped faces in the folder previously mentioned called 'aligned_face'?

import face_recognition
import cv2
import numpy as np

# load image and find face locations.
image = face_recognition.load_image_file("sample.jpg")
face_locations = face_recognition.face_locations(image, model="hog")

# detect 68-landmarks from image. This includes left eye, right eye, lips, eye brows, nose and chins
face_landmarks = face_recognition.face_landmarks(image)

'''
Let's find and angle of the face. First calculate 
the center of left and right eye by using eye landmarks.
'''
leftEyePts = face_landmarks[0]['left_eye']
rightEyePts = face_landmarks[0]['right_eye']

leftEyeCenter = np.array(leftEyePts).mean(axis=0).astype("int")
rightEyeCenter = np.array(rightEyePts).mean(axis=0).astype("int")

leftEyeCenter = (leftEyeCenter[0],leftEyeCenter[1])
rightEyeCenter = (rightEyeCenter[0],rightEyeCenter[1])

# draw the circle at centers and line connecting to them
cv2.circle(image, leftEyeCenter, 2, (255, 0, 0), 10)
cv2.circle(image, rightEyeCenter, 2, (255, 0, 0), 10)
cv2.line(image, leftEyeCenter, rightEyeCenter, (255,0,0), 10)

# find and angle of line by using slop of the line.
dY = rightEyeCenter[1] - leftEyeCenter[1]
dX = rightEyeCenter[0] - leftEyeCenter[0]
angle = np.degrees(np.arctan2(dY, dX))

# to get the face at the center of the image,
# set desired left eye location. Right eye location 
# will be found out by using left eye location.
# this location is in percentage.
desiredLeftEye=(0.35, 0.35)
#Set the croped image(face) size after rotaion.
desiredFaceWidth = 128
desiredFaceHeight = 128

desiredRightEyeX = 1.0 - desiredLeftEye[0]

# determine the scale of the new resulting image by taking
# the ratio of the distance between eyes in the *current*
# image to the ratio of distance between eyes in the
# *desired* image
dist = np.sqrt((dX ** 2) + (dY ** 2))
desiredDist = (desiredRightEyeX - desiredLeftEye[0])
desiredDist *= desiredFaceWidth
scale = desiredDist / dist

# compute center (x, y)-coordinates (i.e., the median point)
# between the two eyes in the input image
eyesCenter = ((leftEyeCenter[0] + rightEyeCenter[0]) // 2,
    (leftEyeCenter[1] + rightEyeCenter[1]) // 2)

# grab the rotation matrix for rotating and scaling the face
M = cv2.getRotationMatrix2D(eyesCenter, angle, scale)

# update the translation component of the matrix
tX = desiredFaceWidth * 0.5
tY = desiredFaceHeight * desiredLeftEye[1]
M[0, 2] += (tX - eyesCenter[0])
M[1, 2] += (tY - eyesCenter[1])

# apply the affine transformation
(w, h) = (desiredFaceWidth, desiredFaceHeight)
(y2,x2,y1,x1) = face_locations[0] 

output = cv2.warpAffine(image, M, (w, h),
    flags=cv2.INTER_CUBIC)

output = cv2.cvtColor(output, cv2.COLOR_BGR2RGB)

this seems not related with opencv ? Cos you already achieved cropping? — Yunus Temurlenk, Sep 01 '20 at 10:12
Is this question really related with the `opencv` ? You want to walk-through subfolders which contain face images, apply the algorithm store it in `aligned_face` folder? — Ahmet, Sep 01 '20 at 10:13

score 0 · Answer 1 · answered Sep 01 '20 at 10:29

Seems like the question is "How do I loop through files and write back"?

Several questions on how to iterate through files: How do I align and crop the images located in subdirectories for face recognition?, Recursive sub folder search and return files in a list python

To crop images you can use Pillow: https://pillow.readthedocs.io/en/stable/

To save images:

f = open('filepath/filename.png','wb') #wb = write byte. Path from the recursive search
f.write(image) #From opencv+numpy->pillow
f.close()

This is not what I was trying to ask but thanks anyways – Shikhar Ghimire Sep 01 '20 at 10:37 — Shikhar Ghimire, Sep 01 '20 at 10:37

score 0 · Answer 2 · answered Sep 02 '20 at 13:49

Firstly, walk all directories and sub directories. The following code block will store the exact paths of images in sub-directories.

employees = []
for r, d, f in os.walk(db_path): # r=root, d=directories, f = files
    for file in f:
        if ('.jpg' in file):
            exact_path = r + "/" + file
            employees.append(exact_path)

employees list stores exact image paths. We need to detect and align faces. Herein, deepface offers detection and alignment in a single function.

#!pip install deepface
from deepface import DeepFace
import cv2

index = 0
for employee in employees:
   aligned_face = DeepFace.detectFace(employee)
   cv2.imwrite('aligned_face/%d.jpg' % (index), aligned_face)
   index = index + 1

This will save detected and aligned faces in aligned_face folder.

Deepface also offers a face recognition module as well but you asked how to detect and align faces.

Besides, you can pass different methods such as opencv, ssd, dlib or mtcnn to detectFace function as an argument. DeepFace.detectFace(employee, detector_backend = 'mtcnn') — johncasey, Sep 06 '20 at 05:27

How do I align and crop the images located in subdirectories for face recognition?

2 Answers2