Thursday 23rd of November 2017
 

Moving Foreground Detection Based On Spatio-temporal Saliency


Yang Xia, Ruimin Hu, Zhongyuan Wang and Tao Lu

Detection of moving foreground in video is very important for many applications, such as visual surveillance, object-based video coding, etc. When objects move with different speeds and under illumination changes, the robustness of moving object detection methods proposed so far is still not satisfactory. In this paper, we use the semantic information to adjust the pixel-wise learning rate adaptively for more robust detection performance, which are obtained by spatial saliency map based on Gaussian mixture model (GMM) in luma space and temporal saliency map obtained by background subtraction. In addition, we design a two-pass background estimation framework, in which the initial estimation is used for temporal saliency estimation, and the other is to detect foreground and update model parameters. The experimental results show that our method can achieve better moving object extraction performance than the existing background subtraction method based on GMM.

Keywords: Moving Object Detection, Background Subtraction, Visual Saliency, Gaussian Mixture Model

Download Full-Text


ABOUT THE AUTHORS

Yang Xia
Yang Xia received the B.S degrees from Wuhan University of technology in 2005, Wuhan, China. He is currently pursuing the Ph.D. degree in National Engineering Research Center For Multimedia Software, Wuhan University, Wuhan, China.His research interests include image/video processing, video coding and artificial intelligence.

Ruimin Hu
Ruimin Hu received the B.S and M.S degrees from Nanjing.University of Posts and Telecommunications, Nanjing China,in 1984 and in 1990 respectively. and Ph.D degree in Communication and Electronic System from Huazhong University of Science and Technology, Wuhan, China in 1994. Dr. Hu is the director of National Engineering Research Center For Multimedia Software, Wuhan University and Key Laboratory of Multimedia Network Communication Engineering in Hubei province. He is Executive Chairman of the Audio Video coding Standard (AVS) workgroup of China in Audio Section. He has published two books and over 100 scientific papers. His research interests include audio/video coding and decoding, video surveillance and multimedia data processing.

Zhongyuan Wang
Zhongyuan Wang received the B.S. degree and M.S degree in computer science from Wuhan University, Wuhan, China, in 1995 and 2001, and he received the Ph.D. degree in Communication and Information System in Wuhan University in 2008. From 2001, he worked as a Member of Research Staff in National Multimedia Software Engineering Research Center of Wuhan University. His research interests include video compression, multimedia communications.

Tao Lu
Tao Lu received the B.S and M.S degrees from Computer Science and Engineering Department, Wuhan institute of technology, Wuhan, China. He is currently pursuing the Ph.D. degree in National Engineering Research Center For Multimedia Software, Wuhan University, Wuhan, China. His research interests include image/video processing, computer vision and artificial intelligence.


IJCSI Published Papers Indexed By:

 

 

 

 
About IJCSI

IJCSI is a refereed open access international journal for scientific papers dealing in all areas of computer science research...

Learn more »
Join Us
FAQs

Read the most frequently asked questions about IJCSI.

Frequently Asked Questions (FAQs) »
Get in touch

Phone: +230 911 5482
Email: info@ijcsi.org

More contact details »