a fast algorithm for tracking human faces based on chromatic histograms pattern recognition letters,...

29
A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Lette rs, 1999 Speak: M. Q. Jing 4/23/2001 國國國國國國 國國國國國國國 lab

Upload: aleesha-fitzgerald

Post on 17-Jan-2016

216 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

A fast algorithm for tracking human faces based on chromatic histograms

Pattern Recognition Letters, 1999

Speak: M. Q. Jing4/23/2001

國立交通大學 自動化資訊處理 lab

Page 2: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

Outline Introduction Chromatic histogram operations

Creation of a chromatic histogram Backprojection of a chromatic histogram

Face tracking algorithm Localization of a face region Tracking the face in the sequences

Experimental results

Page 3: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

Introduction The solutions on motion

Motion Estimate (Haralick, Horn, Desilva) Precise approximation of the motion [DrawBack] heavy computation

Color histogram approach Insensitive to rotation,scaling,deformation Immune to the noises and cammera’s small

changing.

Page 4: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

Chromatic histogram operations

Color Model: HSI Color Model

2/12

1

)])(()[(

)]()[(21

cos

)],,[min()(

31

)(3

1 I

HSI RGB

bgbrgr

brgrH

bgrbgr

S

bgr

Page 5: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

Why do we use HSI

The chromaticity and intensity is separated Reduce the effect of illumination

Reduce the dimension of histogram from 3 to 2 Speedup the process

Page 6: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

Creation of a chromatic histogram

Each of the H and S axes are quantized into 32 levels.

SaturationHue

sample(x,y)=(R,G,B)-> map to (h,s)->Quantized->Histogram

Page 7: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

Create the histogram Create the face model to get the

skin color histogram

SaturationHue

Page 8: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

Backprojection of a chromatic histogram

SaturationHue

SaturationHue

Model

Test

1

1.which bin 2.get value 2

Page 9: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

Backprojection of a chromatic histogram

Algorithm: Step 1: bx,y=M h(Cxy) , Step 2: Convolving bx,y with a blurring mask

Where h(Cxy)= the bin corresponding to Cxy, M I= the histogram of the Model with ith bin.

Page 10: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

More example

Page 11: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

Face tracking algorithm How to find a face in the initial

frame Face region lies within a color range

Face region

Historgram foreach region

Page 12: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

Face tracking algorithm Compute an average of the face histor

grams

Face model histogram F (100 faces histogram)

Page 13: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

Face tracking algorithm

1

2

3

Steps

1. Backproject2. Binarized & CC3. Search a ellipse

Page 14: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

Finding an Ellipse An Ellipse which best fits the

connected component is computed.

CCy)(x,

iji,

2,00,21,1

00010010

),(ij

)()x-(x moment central

2/)]/()arctan[(2n orientatio

/,/mx :)y,x(center

),(m :monent

j

CCyx

ji

yy

mmym

yxfyx

Page 15: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

Least-Squares Curve Fitting Length of major (a) and minor (b)

axis:

8/1

),(

2

3

),(

2

4/1

8/1

),(

2

3

),(

2

4/1

]cos)(sin)[(

]sin)(cos)[(

)/4(

]sin)(cos)[(

]cos)(sin)[(

)/4(

CCyx

CCyx

CCyx

CCyx

yyxx

yyxx

b

yyxx

yyxx

a

Proof: computer and robot vision I, page 623

Page 16: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

Finding an Ellipse The golden ratio of ellipse is picked

up.

)1987 (Farkas,

)2/5(1face ideal afor ratioGolden

Page 17: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

Tracking the face region Step 1: a new face model histogram F is constructed fr

om the detected face. More precise face model, because tracking the s

ame face.

Step 2: Backprojected onto the next frame.

Step 3: An elliptical mask is used for searching No ellipse finding,saving the computation cost

why

why

Page 18: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

Tracking the face region Step 4: compute the sum of the

values of all pixels within the elliptical mask.

Step 5: return maximum response location

Page 19: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

Tracking the face region Prevent the searching cost for (left->right) & ( top -> down) Using motion information :

(Xi+1,Yi+1)=(2Xi - Xi-1 , 2Yi - Yi-1)

Xi-1 ,Yi-1Xi,Yi

(Xi+1,Yi+1)

Page 20: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

Experements UperSPARC RISC with 60MHz, 64 MB Real-time processing

7 frames/sec (160x120) 3.5 frames/sec (240x180)

Page 21: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab
Page 22: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

Face tracking (small face)

Page 23: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

Face tracking (large face)

error

Page 24: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab
Page 25: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

Face tracking results using skip factor 5

change

error

Page 26: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

changeerror

Page 27: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

Conclusion A histogram backprojection only need

s a simple replacement operation Insensitive to small deformation and

occlusion Because we use color information

Two feature are used Face shape & chromatic

Page 28: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

Conclusion It cannot handle non-forward faces

Because we use a ellipse model to find a face.

Zoom-in and Zoom-out We fixed the ellipse size due to

reducing the computing cost.

Page 29: A fast algorithm for tracking human faces based on chromatic histograms Pattern Recognition Letters, 1999 Speak: M. Q. Jing 4/23/2001 國立交通大學 自動化資訊處理 lab

Analysis 因為 Tracking algorithm 在第一張人臉

抓取後 , 即 update face model histogram,

=> 所以第一張的人臉一定要抓的準確 ,否則將造成一系列的錯誤

Face color histogram 是假設大家的膚色類似 , 但是若是 testing 有黑人 , 白人 ,則會造成 histogram 分佈加大 , 使得 backprojected 圖形更難處理 .