Full text: Real-time imaging and dynamic analysis

     
   
   
  
     
        
       
   
   
    
    
    
    
     
   
   
   
    
  
  
  
   
   
    
    
   
   
   
  
     
  
   
   
    
   
380 1 1 1 1 
  
   
  
  
  
  
  
  
  
  
EASURED P13 Y —— 
3707 $. $, EXP TED P13 Yy —— ' 
^. | 260} 4 i | à | | | $ ] 1 
& 350 uw v Ui + tA | | 1 b $Y à | A P S | 
3 S40 hj)! ib Mi? f JP VP y be 43 114 7 
> $$] » i M 3 | = dé d $4 N d & 
330 || ¢ si i yo? ww | 
320 ¥ ] 
310 i i i 
0 50 100 150 200 250 
Frame Number 
(a) 
400 T T T T T T T 
i» % a 
350 F A 
= 300} = 3 
> a 
P 250-5. e - 
5 a % 
> 200 } | 
150 | de + Ë 
100 r L 1 1 L L 1 
100 150 200 250 300 350 400 450 500 
X-axis (pixel) 
(c) 
lational motion. Distortion shown in the experiment 
is caused by the insufficiency. However, collecting 
enough sample images requires subject's effort. 
Our future work will concentrate on the reduction 
of the distortion caused by insufficient orthonormal 
basis. 
5. REFERENCES 
[1] Masashi Setaka, Time-series analysis of multi- 
point movements of lips and adjacent regions 
during mastication, Hokkaido Journal of Dental 
Science, 15, pp.89-107, 1994 ( in Japanese ) 
[2] M. Covell, Eigen-points: Control-point Location 
using Principal Component Analysis, 2nd Int. 
  
  
  
  
  
  
  
  
  
400 T T TT — 
x = 
350 | 
> 300 + + 
X tQ 
£ - 2 
0 250 | "e 
x S 
9 ve 
> : 200} A 
150 } v A 
100 
100 150 200 250 300 350 400 450 500 
X-axis (pixel) 
(b) 
400 T T T T 
x - 75 
350 | 
= 300} 2 
X "s 
2 - 
" 250 | s 
i 
> 200} S vi 
150 | 4 id 
7 
100 1 1 1 L 1 1 
100 150 200 250 300 350 400 450 500 
X-axis (pixel) 
(d) 
Figure 6: (a)Vertical motion of the point C2. Cross: estimation. Diamond: true position. The amplitude of 
the expected vertical motion is smaller than actual motion. This is probably caused by in sufficient span of 
the basis. Figures (b)-(d) are the trajectories of markers. Figure (b) is the output of tracker (actual motion). 
Figure (c) is the estimated trajectory for the training sample, and (d) is the result for an test image. 
Conf. on Automatic Face and Gesture Recogni- 
tion, 1996 
  
q) 
KEY WC 
ABSTRA 
Ultrasono 
noninvasi 
research « 
tool for t! 
images a 
photograr 
data accu 
Each year : 
ischemic str 
contribute 1 
Ultrasonogra 
noninvasive 
a significant 
identificatior 
surgical trea 
other either 
provide 2-D 
However, | 
certainly a p 
dimensional 
abnormal fo: 
alters the ge 
complex we 
Across this 
parameters i 
general been 
delineate th 
contribute si 
more efficac 
stroke that v 
Stroke is cu 
cause of m 
among the
	        
Waiting...

Note to user

Dear user,

In response to current developments in the web technology used by the Goobi viewer, the software no longer supports your browser.

Please use one of the following browsers to display this page correctly.

Thank you.