GMDN: A lightweight graph-based mixture density network for 3D human pose regression

2021 ◽  
Vol 95 ◽  
pp. 115-122 ◽  
Author(s):  
Lu Zou ◽  
Zhangjin Huang ◽  
Naijie Gu ◽  
Fangjun Wang ◽  
Zhouwang Yang ◽  
...  
Keyword(s):  
2010 ◽  
Vol 32 (3) ◽  
pp. 750-754 ◽  
Author(s):  
Ying-na Deng ◽  
Hong Zhu ◽  
Wei Liu ◽  
Xiao-dan Zhang

2011 ◽  
Vol 33 (6) ◽  
pp. 1413-1419
Author(s):  
Yan-chao Su ◽  
Hai-zhou Ai ◽  
Shi-hong Lao

1995 ◽  
Vol 60 (8) ◽  
pp. 1274-1280 ◽  
Author(s):  
Kamil Wichterle

Analysis of extended data on turbine impeller power input in geometrically similar agitated baffled tanks shows that the power number Po is a function of Reynolds number Po = Po*(Re) until the emergence of surface aeration. Though it is usually anticipated that Po* = const in high Reynolds number region, some, whatever weak, function should be taken into consideration in more detailed analysis of the power data even here. In practice, disturbances of level and gas captured in the impeller region play also a significant role, namely in smaller tanks at higher impeller speeds. Decrease of power input can be explained by decrease of gas-liquid mixture density, or in other words by increase of efficient gas holdup eE just in the impeller region. The value eE defined by the relation Po = Po*(Re)/(1 + eE) was determined from the available data. Like other effects of the surface aeration it depends mainly on the dimensionless number Nc = (We Fr)1/4. A simple correlation eE (Nc) is suggested as a correction factor for prediction of impeller power in presence of gas capture.


Author(s):  
Jinbao Wang ◽  
Shujie Tan ◽  
Xiantong Zhen ◽  
Shuo Xu ◽  
Feng Zheng ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document