2. What is Face Detection?
Given an image, tell whether there is any human face
or not.
If there is then find the location and size of each
human face in the image.
Classification between face and non-face.
Results for Training_1.jpg
3. Difference between Face
Detection and Face
RecognitionFD:-Only two classifications
face or non face
FR:- have multiple classifications, adjusted by the
number of individuals who want to be recognized.
One person vs. all the others
Classification from face shape, of form eyes, nose,
mouth, etc.
FR process requires the first FD
4. Why Face Detection is
Important?
First step for any automatic face recognition system.
First step in many Human-Computer Interaction
systems and Man-machine Interaction.
• Expression Recognition
• Cognitive State/Emotional State Recogntion
First step in many surveillance systems
Tracking: Face is a highly non rigid object.
A step towards Automatic Target Recognition(ATR) or
generic object detection/recognition.
5. Methods for Face Detection
Knowledge-based methods:
Encode what constitutes a typical face.
e.g., the relationship between facial features
Feature invariant approaches:
Aim to find structure features of a face that exist even when
pose, viewpoint or lighting conditions vary
Template Matching:
Several standard patterns stored to describe the face as a whole
or the facial features separately
Appearance-based methods:
The models are learned from a set of training images that
capture the representative variability of faces.
1 2 3
6. System Architecture For Face
Detection
Input RGB
JPEG image
Average
Face
Skin
Segmentation
RGB to YCbCr
RGB to HSV
Threshold to
determine skin
regions
Multi-
resolution
Iterative
Template
Matching
Classifier
Skin
Pixels
Face/
Non-Face
7. Flowchart for Face
detection
Input Image
Find Skin
Regions
Count number of
holes
Detected Face
region
Template
Matching
Skin
Segmention
Non-FaceHoles
>=1
8. Process of finding Skin pixels in image.
Reject as much “non-skin” of the image as possible.
Detects Skin region in images and remove background
parts using threshold.
Remove other body parts by applying binary image
processing.
Skin Segmentation and
Thresholding
9. Template Matching
A Technique for finding small parts of an image
which match a template image.
Used to detect the more accurate faces, and
neglect the non-faces
Determine the face locations such as based on
the correlation values.
Resize and rotate template image as
requirement for finding human face in image.
11. Template face Rotated Template
face
1. A skin region 2. skin region
without holes
3. Template face is located in the
center of the skin region
12. 5. As in (4), but inverted4. As in figure (3) but with a
hole in the template face
6. Previous image is multipled
by original one
As in (6), but adding the
Template face to it.
8. Final Result
13. Face Detection : Current
State
Front-view face detection can be done at more than
15 frames per second on 320x240 black-and-white
images on a 700MHz PC with ~95% accuracy.
Detection of faces is faster than detection of edges!
Side view face detection remains to be difficult.
14. Face Detection : Challenges
Templates needs to be initialized near the face images.
Difficult to enumerate templates for different poses
Out-of-Plane Rotation: frontal, 45 degree, profile, upside
down
Facial Expressions
In-Plane Rotation
Image conditions:
• Size
• Lighting condition
• Distortion
• Noise
• Compression