close

Вход

Забыли?

вход по аккаунту

?

Color Segmentation

код для вставкиСкачать
Color Segmentation
• View the YIQ color space:
-Y=luminance, I=hue, Q=saturation
• Human skin occupy a small portion of the I and Q
spaces.
• From training images, compare and contrast hue
and saturation of:
faces only vs. entire image
Hue and Saturation
Histogram of Q Components of Training7.jpg
5
14
x 10
12
10
8
6
4
2
0
-150
-100
-50
0
50
100
150
Faces
Training Image
Q Distribution
Mask After Color Segmentation
• Skin elements remain.
• Holes in faces later eliminated with hole-filling
Mask After Object Removal
Based on size distribution of remaining objects, remove small ones
Correlation Template Matching
I – Average Face
• First attempt – Average face
• Taking average of all faces from ground truth
masks
H пЂЅ
1
N
N
x
i
i пЂЅ1
• Results – Less than satisfactory.
– Face with distinguishing features blurred
– Correlation separation is not high, identifies many skin
color regions (clothing, background) as false positives.
Correlation Template Matching
II – Edge detection
• After color segmentation, most remaining regions
are composed of skin-color tones.
• Distinguishing features resides in edges
– Use Canny edge filter on black-white images for extraction
– Composed average face using edges, scaled to mean zero
Correlation comparison
• Average face template
– Poor separation between faces
– Difficult to identify face centroid
• Edge face template
– Better separation between faces
– Peaks (centroid) more easily
identifiable
Region counting - Supplementary method
• The edge outlines have clearly identifiable connected regions
• Can be counted, and statistics used to help reject clutter
Number of regions: 14
Number of regions: 43
Detection Algorithm
– Correlation – Degree of matching
– Dimensions – height, width
– Region counting – complexity of image
Single face
Correlation
Dimensions
Region counting
Dimensions
Region counting
Multiple faces
Correlation
Multi-face
detection
Multiple Faces within a Single
Region
• Search for peaks in
correlation
• A single face may give
multiple peaks
• Estimate expected
number of faces
within Region
• Do not want repeats
Find Largest Peak
• Find largest peak in
correlation
• Location of first peak
• Exclude area of radius R
(about peak) from rest of
search
• R determined dynamically
from size of region and
number of expected faces
Next Peak
• Find next largest peak
• Exclude area (of radius R)
surrounding both peaks
from further search
• Continue search in this
manner until desired
number of peaks found
Find Multiple Faces
• Stop search if there are
no more peaks to be
found
(Number of peaks found
can be fewer than
estimate)
• Each peak location
corresponds to face
center location
Conclusion
• Reasonably successful performance
– Misses
– False positives/repeats
• Algorithm relies heavily on Color Segmentation
and Edge Extraction
• Difficulty with closely-spaced faces
– Separation
– Detecting multiple faces in single region (correct
estimate)
Face Detection
Gender Recognition
1
2 (19)
1 (1)
2
1 (20)
2 (0)
3
3 (18)
2 (0)
4
3 (18)
2 (0)
5
5 (13)
2 (0)
6
7
8
9
10
11
12
13
14
15
16
17
Документ
Категория
Презентации
Просмотров
5
Размер файла
1 280 Кб
Теги
1/--страниц
Пожаловаться на содержимое документа