DocumentCode
2214049
Title
Application of Artificial Neural Network in Video Compression Coding
Author
Zhang, Yong Zhang ; Zhang, Ming Ming
Author_Institution
Sch. of Control Sci. & Eng., Univ. of Jinan, Jinan
Volume
1
fYear
2008
fDate
19-21 Dec. 2008
Firstpage
207
Lastpage
210
Abstract
The goal of video compression is to reduce video data rate on the premise of visual effect as much as possible. During the compression of image, to get a high compressed ratio,people always only get a degraded image. This article aims to estimate the degraded model of the video compresed image through using the neural network and H.264 rate-distortion model.we can improve the quality of the compressed image, using neural network´s predictive ability strongly and high fault-tolerance, improving the quantization parameter prediction in the video encoding, completing video image reconstruction. The experiments show that the neural network can improve the image quality of video compressed image greatly.
Keywords
artificial intelligence; data compression; encoding; fault tolerance; image representation; neural nets; quantisation (signal); video coding; artificial neural network; fault-tolerance; quantization parameter prediction; video compression; video compression coding; video encoding; video image reconstruction.The; Artificial neural networks; Degradation; Fault tolerance; Image coding; Neural networks; Predictive models; Quantization; Rate-distortion; Video compression; Visual effects; RBF neural network; rate-distortion; vector quantization; video compression;
fLanguage
English
Publisher
ieee
Conference_Titel
Information Management, Innovation Management and Industrial Engineering, 2008. ICIII '08. International Conference on
Conference_Location
Taipei
Print_ISBN
978-0-7695-3435-0
Type
conf
DOI
10.1109/ICIII.2008.298
Filename
4737529
Link To Document