Despite the natural synergy among image understanding and image processing research on computer vision (e.g., video surveillance) and visual information compression (e.g., image/videocoding) has been carried out mostly in isolation. For high performance we need to integrate the knowledge from computer vision and video coding technology. This need arises due to the recentexpansion of new technologies based on the wireless networks in the burgeoning Internet, closecircuits security cameras, mobile devices, home entertainment, and YouTube. While wirednetworks today enjoy almost unbounded bandwidth, wireless networks still operate under stringent bandwidth constrains. Thus, video communication over wireless networks requiresimprovement in the areas of compression, quality, and computational complexity in the video coding technology. In this chapter, we will cover these areas based on recent research. Firstly wepresent dynamic background modeling techniques to generate a dynamic frame popularly known as McFIS (the most common frame of a scene) from dynamically challenging environments for detecting moving objects, secondly we present a number of advanced video coding techniques for improving rate-distortion performance as well as reducing computational complexity compared to the state-of-the-arts methods, and finally we present computer vision aided video coding technique by combining computer vision technique (such as dynamic background) and video compression technique to improve the coding performance in term of quality (subjective and objective), compression, and computational complexity for stringent bandwidth carrier such as wireless network.
|Title of host publication||Advanced video communications over wireless networks|
|Editors||Ce Zhu, Yuenan Li|
|Place of Publication||Boca Raton, FL|
|Number of pages||28|
|Publication status||Published - 2013|