Abstract— Inter-camera registration in multi-view systems with overlapped views has an especially long and sophisticated research history within the computer vision community. It represents a genuine challenge as a result of necessary data at decoder for generating the side information without the a priori knowledge of every instant camera position. This paper proposes a remedy to the problem centered on successive multi-view registration and motion compensated extrapolation for on-the-fly re-correlation of two views at decoder. This novel technique for side information generation is camera position independent, robust and flexible with regard to any free localization of the cameras. Furthermore, it doesn't require any additional information from encoders nor communication between cameras or offline training stage. Link between simulation demonstrate significant data compression in Wireless Sensor Networks, during keeping data quality.
Index Terms— Distributed video coding, image mosaicing, side information generation.
M. Haqqani is with the Computer Engineering Department, Shiraz University, Shiraz, Iran (e-mail: mohammad.haqqani@gmail.com).
K. Raahemifar is with the Electrical and Computer Eng. Department, Ryerson University, Toronto, Canada (e-mail: kraahemi@ee.ryerson.ca).
M. Fathy is with the Computer Engineering Department, Iran University of Science and Technology, Tehran, Iran (e-mail: mahfathy@iust.ac.ir).
[PDF]
Cite: Mohammad Haqqani, Kaamran Raahemifar, and Mahmood Fathy, " A Novel Side Information Generation Algorithm in Distributed Multi-View Video Coding," International Journal of Information and Electronics Engineering vol. 4, no. 4, pp. 279-283, 2014.