.png){kind=icon}
.png){kind=icon}
Return
Arbitrarily shaped virtual-object based video compression
.png){kind=icon}
OA
AI
DOI：10.1007/s11042-011-0869-z.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Object based compression techniques are widely believed to have the potential to give the best compression results for a given signal quality. However, true object tracking and extraction are difficult and computationally expensive. In this paper, an arbitrarily shaped virtual-object compression method is developed. The method is similar to the object based compression methods in that it separates the changing portion of the video from the stationary portion, and encodes them independently. The changing portion of the video is grouped as a 3D arbitrarily shaped virtual-object whereas the unchanged portion of the video is grouped as background. The arbitrarily shaped virtual object is coded using 3D wavelet compression whereas stationary background is coded as a single frame using 2D wavelet compression. Experimental results demonstrate that the newly developed method has comparable performance with the state-of-the-art compression methods and significantly outperforms rectangular virtual-object compression.
Keywords:
Arbitrarily shaped virtual-object
Shape adaptive wavelet transform
3D wavelet compression
Journal
IF:
3
Papers:
1.9W
Citations:
3.2W