Abstract

A new perceptual video coder is proposed here in which the coding structure is coupled with a vision model to produce a coded sequence with an improved quality for the same compression ratios when compared to the existing standards. The performance of the proposed coder is compared with that of the standard codec and it gives a very good perceptual quality especially at lower bit rates. Recently, there are two interesting trends in image and video coding research. One is to use Human Visual System (HVS) models to improve the current state-of-the-art coding algorithms by better exploiting the properties of the intended receiver with the help of Contrast Thresholding and Masking. The other is to design a Foveation based visual sensitivity model. In this work these two trends are followed and it is proposed that video coding based on model-1 supplies good quality-compression performance. Under the assumption that human visual perception is highly adapted for extracting structural information from a scene, an alternative framework for quality assessment based on the degradation of structural information is used to assess the perceptual quality of the compressed images.

 Keywords: video compression, Human Vision System, Contrast Sensitivity, Foveation, Discrete Cosine Transform.

(P1150827001, 561 KB)

Biographies:

	Sreelekha G received her Bachelors Degree in Electronics Communication and Engineering from College of Engineering Trivandrum, Kerala, India in 1997 and her Masters in Digital Electronics from Cochin University of Science and Technology, Kerala, India. She is Senior Lecturer in the Department of Electronics and Communication Engineering, National Institute of Technology Calicut, Kerala, India. She is currently doing her research in the area of perceptual image coding. Her areas of interest are Digital signal processing, Filter banks, Wavelet theory and Image & Video processing.
	T. Rajesh earned his Bachelors Degree in Electronics Communication and Engineering from Jayaprakash Narayan College of Engineering, Andhra Pradesh, India and his Masters in Electronics Design and Technology from National Institute of Technology, Kerala, India. His areas of interest are Digital signal processing, Image and Video processing.
	P.S. Sathidevi received B.Tech Degree in Electronics Engineering from Regional Engineering College, Calicut (Calicut University) in 1985, M.Tech Degree in Electronics from Cochin University of Science and Technology, Cochin in 1987 and Ph.D from Regional Engineering College Calicut (Calicut University) in 2003 in the field of Speech and Audio Processing. She has been working as Lecturer in the Dept. of Electronics Engineering, National Institute of Technology Calicut (Formerly REC Calicut) from 1990 to 2003, as Asst. Professor from 2003 to 2007 and as Professor from 2007 onwards. Her current interests include Perceptual audio coding, Image & Video Coding, Speech Coding, and Speech Enhancement.

BibTex:

@ARTICLE{F1150827001,

AUTHOR = {G. Sreelekha and T. Rajesh and P.S. Sathidevi},

TITLE = {A NEW PERCEPTUAL VIDEO CODER INCORPORATING HUMAN VISUAL SYSTEM MODEL},

JOURNAL ={ICGST International Journal on Graphics, Vision and Image Processing, GVIP},

YEAR = {2008},

VOLUME = {08},

PAGES={7--15},

ISSUE ={III}

}

(P1150827001, 561 KB)