Face detection in real time with MTCNN ? 'NoneType' object has no attribute 'size'

Part 1 (2019)
1

i’m trying to build face recognition in real time in neural network(facenet network) using pytorch and face detection using MTCNN i’ve tried this but doesnt work

import cv2

capture = cv2.VideoCapture(0)

while(True):

    ret, frame = capture.read()
    frames_tracked = []

    print('\rTracking frame: {}'.format(i + 1), end='')
    boxes,_ = mtcnn.detect(frame)
    frame_draw = frame.copy()
    draw = ImageDraw.Draw(frame_draw)
    for box in boxes:
        draw.rectangle(box.tolist(), outline=(255, 0, 0), width=6)

       frames_tracked.append(frame_draw.resize((640, 360), Image.BILINEAR))

    d = display.display(frames_tracked[0], display_id=True)
    i = 1
    try:
       while True:
           d.update(frames_tracked[i % len(frames_tracked)])
           i += 1
    except KeyboardInterrupt:
        pass

  if cv2.waitKey('q') == 27:
     break

 capture.release()
 cv2.destroyAllWindows()

AttributeError: ‘NoneType’ object has no attribute ‘size’ thanks for any advice

2

The full error message would be more useful. We would know what part of the code caused the error.

3
4

Can you post the full error not just the last line of it so we can tell what line the error is coming from.
For example (it is usually longer):

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "<stdin>", line 3, in a
ValueError
5

if it detect a face in the first frame it will raise this error

ValueError Traceback (most recent call last)
in
13 frame_draw = frame_pil.copy()
14 draw = ImageDraw.Draw(frame_draw)
—> 15 if boxes:
16 for box in boxes:
17 draw.rectangle(box.tolist(), outline=(255, 0, 0), width=6)

ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()
if there wasnt a face in the first frame it will raise this error

IndexError Traceback (most recent call last)
in
18 frames_tracked.append(frame_draw.resize((640, 360), Image.BILINEAR))
19 clear_output(wait=True)
—> 20 d = display.display(frames_tracked[0], display_id=True)
21 i = 1
22 try:

IndexError: list index out of range

6 
import cv2

capture = cv2.VideoCapture(0)

while not cv2.waitKey('q') == 27:

    ret, frame = capture.read()
    frames_tracked = []

    print('\rTracking frame: {}'.format(i + 1), end='')
    boxes,_ = mtcnn.detect(frame)
    frame_draw = frame.copy()
    draw = ImageDraw.Draw(frame_draw)

    for box in boxes:
        draw.rectangle(box.tolist(), outline=(255, 0, 0), width=6)
        frames_tracked.append(frame_draw.resize((640, 360), Image.BILINEAR))

    if frames_tracked:
        d = display.display(frames_tracked[0], display_id=True)
        i = 1
        try:
           while True:
               d.update(frames_tracked[i % len(frames_tracked)])
               i += 1
        except KeyboardInterrupt:
            pass

 capture.release()
 cv2.destroyAllWindows()

Does this behave differently?

7

TypeError Traceback (most recent call last)
in
3 capture = cv2.VideoCapture(0)
4
----> 5 while not cv2.waitKey(‘q’) == 27:
6
7 ret, frame = capture.read()

TypeError: an integer is required (got type str)
no , does it work for you

8

I don’t have mtcnn so I cannot test it but that error can be fixed simply:
while not cv2.waitKey(33) == ord('q'):

I will install it…

I can run it now and I am getting some errors.

1 Like
9

thank you for your helping

1 Like
10

I can’t even get an image to save yet. Did you manage to write something that can capture an image and just show it?

11

install mmcv

video = mmcv.VideoReader('video.mp4')

frames = [Image.fromarray(cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)) for frame in video]

display.Video(‘video.mp4’, width=640)

frames_tracked = []
for i, frame in enumerate(frames):
print(’\rTracking frame: {}’.format(i + 1), end=’’)

# Detect faces
boxes, _ = mtcnn.detect(frame)

# Draw faces
frame_draw = frame.copy()
draw = ImageDraw.Draw(frame_draw)
for box in boxes:
    draw.rectangle(box.tolist(), outline=(255, 0, 0), width=6)

# Add to frame list
frames_tracked.append(frame_draw.resize((640, 360), Image.BILINEAR))

print(’\nDone’)

13

ValueError Traceback (most recent call last)
in
14
15 img = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
—> 16 boxes,_ = detector.detect_faces(img)
17 frame_draw = frame.copy()
18 draw = ImageDraw.Draw(frame_draw)

ValueError: not enough values to unpack (expected 2, got 0)

14

I am getting the same error. I think we just have to do:
boxes = detector.detect_faces(img)

16

empty list

19

Finally here we go!!! :grinning:

import cv2
from mtcnn import MTCNN
from PIL import Image, ImageDraw
import os
import numpy as np
import time

os.environ['KMP_DUPLICATE_LIB_OK']='True'

detector = MTCNN()
cap = cv2.VideoCapture(0)

while True:
    ret, frame = cap.read()

    if ret:
        boxes = detector.detect_faces(frame)
        frame_draw = frame.copy()
        img = Image.fromarray(frame_draw)
        draw = ImageDraw.Draw(img)

        for box in boxes:
            draw.rectangle(box["box"], outline=(255, 0, 0), width=6)

        opencvImage = cv2.cvtColor(np.array(img)[..., ::-1], cv2.COLOR_RGB2BGR)
        opencvImage = cv2.resize(opencvImage, (720, 480))
        cv2.imshow('output', opencvImage)

        if cv2.waitKey(1) &0xFF == ord('q'):
            break

    else:
        break

cap.release()
cv2.destroyAllWindows()
1 Like
20

thank you so much
much appreciate

1 Like
21

hi, I need help with multilabel classification using celebA. Can someone help me here?