An Attention Based Spatial Adaptation Scheme for H.264 Videos on Mobiles

Author(s):  
Yi Wang ◽  
Xin Fan ◽  
Houqiang Li ◽  
Zhengkai Liu ◽  
Mingjing Li
2021 ◽  
Vol 0 (0) ◽  
Author(s):  
Christoph Kandlbinder-Paret ◽  
Alice Fischerauer ◽  
Gerhard Fischerauer

Abstract In electrical capacitance tomography (ECT), the resolution of the reconstructed permittivity distribution improves with the number of electrodes used whereas the number of capacitance measurements and the measurement time increases with the number of electrodes. To cope with this tradeoff, we present a phantom-dependent adaptation scheme in which coarse measurements are performed with terminal electrodes interconnected to form a synthetic electrode ring with fewer but larger electrodes. The concept was tested by observing the sloshing of water inside a pipe. We compare the reconstructed results based on eight synthetic electrodes, on 16 elementary electrodes, and on the adaptation scheme involving both the eight synthetic electrodes and some of the elementary capacitances. The reconstruction used the projected Landweber algorithm for capacitances determined by a finite-element simulation and for measured capacitances. The results contain artefacts attributed to the influence of the high permittivity of water compared to the low permittivity of the pipe wall. The adaptation scheme leads to nearly the same information as a full measurement of all 120 elementary capacitances but only requires the measurement of 30 % fewer capacitances. By detecting the fill level using a tomometric method, it can be determined within an uncertainty of 5 % FS.


2006 ◽  
Vol 15 (10) ◽  
pp. 2866-2878 ◽  
Author(s):  
C. Kervrann ◽  
J. Boulanger

Author(s):  
Andrew Macintosh ◽  
Jan McDonald ◽  
Anita Foerster

2005 ◽  
Vol 194 (48-49) ◽  
pp. 5019-5050 ◽  
Author(s):  
Yannis Kallinderis ◽  
Christos Kavouklis

2013 ◽  
Vol 05 (03) ◽  
pp. 36-41 ◽  
Author(s):  
Zi Teng ◽  
Jun Wu ◽  
Min Wang ◽  
Lifeng Su

Sign in / Sign up

Export Citation Format

Share Document