Scalable video coding method, encoder and computer program

摘要

A scalable video coding method for coding a video sequence (10) results in a bit stream (20} partitionable into two layer representations (201, 2O2) of the video sequence (10). The second layer representation (2O2) has an enhanced resolution compared to the first layer representation (2O1.). The method includes generating (si) predictive coding parameters for the second layer using previously reconstructed pixels (14) in the second layer. The method further includes performing (s2) prediction for the first layer, using the predictive coding parameters and the previously reconstructed pixels (14) in the second layer, to generate a first layer prediction signal. The method further includes generating (s3) a first layer residual, based on the difference between the original image and the first layer prediction signal. The predictive coding parameters and the first layer residual are coded (s4) as the first layer representation (2O1). An encoder (30) and computer program are also provided.

主权项

1. A scalable video coding method for encoding a video sequence into a bit stream, the resulting bit stream being partitionable into at least a first layer representation and a second layer representation of the video sequence, wherein the second layer representation has an enhanced spatial or quality resolution compared to the first layer representation, comprising:
generating predictive coding parameters for the second layer representation of at least one region of an image of the video sequence, using previously reconstructed pixels in the reconstructed second layer representation; performing prediction for the first layer representation of the at least one region of the image of the video sequence, using the predictive coding parameters and the previously reconstructed pixels in the reconstructed second layer representation, to generate a first layer prediction signal; generating a first layer residual based on the difference between the at least one region of the image of the video sequence and the first layer prediction signal; and encoding, as at least part of the first layer representation, the predictive coding parameters and the first layer residual.