| 摘要 |
An apparatus and method for scalable video coding is disclosed. The apparatus and method can be used in scalable video coding, where the encoded bitstream needs to provide for spatial and temporal scalability, without transcoding or re-encoding the bitstream. In the prior arts, encoding and decoding are performed such that "full spatial-temporal scalability" is provided. The problem is that not all of the spatial- temporal operating points will be utilized. Catering for these unnecessary operating points penalize the coding efficiency and increase the computation for both encoder and decoder. The present invention encodes and decodes the video data with only "partial spatial-temporal scalability" by performing adaptive spatial-temporal transformations. The algorithm in the present invention is able to encode and decode with minimum "partial spatial-temporal scalability" that satisfy the pre-determined desired operating points. Further spatial-temporal transformations to improve coding efficiency are optional, resulting in a flexible encoding and decoding scheme.
|
