Title
Fixed-point Gaussian Mixture Model for analysis-friendly surveillance video coding
Abstract
•The fGMM is mathematically derived to eliminate the floating-point calculations.•A division simulation algorithm is proposed to replace the division operations.•The fGMM is implemented in FPGA to process HD videos in real-time.
Year
DOI
Venue
2016
10.1016/j.cviu.2015.09.006
Computer Vision and Image Understanding
Keywords
Field
DocType
Surveillance video,Video coding,Analysis-friendly,Background modeling,Fixed-point Gaussian Mixture Model
Computer science,Smart camera,Real-time computing,Coding (social sciences),Artificial intelligence,Codec,System bus,Computer vision,Object detection,Parametric model,Field-programmable gate array,Mixture model,Machine learning
Journal
Volume
Issue
ISSN
142
C
1077-3142
Citations 
PageRank 
References 
4
0.39
28
Authors
4
Name
Order
Citations
PageRank
Wei Chen150.73
Yonghong Tian21057102.81
Yaowei Wang313429.62
Tiejun Huang41281120.48