Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 16

Group Members

 Syed Taimoor Tariq 2018-EE-121

 Saad Ahmad 2018-EE-140

 Arooj Irfan 2018-EE-152

 Mustafa Elfatih 2018-EE-239


Intelligent Student Attendance and
Monitoring System
1 Testing Data

2 Different Algorithms

Content 3 Haar Cascade Classifier

4 Mediapipe Deep Learning

5 Advantages And disadvantages


It is a machine learning
based approach where a
cascade function is trained
Haar Cascade
from a lot of positive
(images with face) and
negative images (images
Classifier
without face).
• Haar Feature Selection
• Creating an Integral Image
• Adaboost used to classify
different features.
• Cascading Classifiers
Face Detection
Detect Multi_scale
Testing on Real Images
Testing on Real Images
Advantages and Disadvantages

 Each OpenCV face detection classifier has its own pros and cons but the major differences are in accuracy and
speed.

Haar_cascade
has following Computationally
Advantages complex and slow
High Detection
and Disadvantages Less robust to occlusion
Accuracy

Low false positive rate Limitations in difficult


lightening conditions
Mediapipe Deep Learning based
Face Detection

 lightweight and highly accurate face detector

 The detector provided by Mediapipe is capable of running at a speed of 200-


1000+ FPS on flagship devices.

 It can detect the non-frontal and occluded faces but fails to accurately
detect the key points in such scenarios.

 By using the short-range model, one can easily ignore the faces in the
background, which is normally required in most of the applications out there,
like face gesture recognition.
Face Detection
Face Detection(short Range model)
Face Detection
Face Detection(long Range model)
Face Detection
Advantages and Disadvantages

 Each OpenCV face detection classifier has its own pros and cons but the major differences are in accuracy and
speed.

Mediapipe
has following
Advantages Highly accurate face Fails to accurately
and Disadvantages detector detect the key points

Unable to detect for


Very Fast Method faces above 5 meters
from the camera

You might also like