| 摘要 |
Methods, systems, and computer program products are provided to provide transform coefficient scaling at a block level in both a video/image encoder and a video/image decoder, rather than at a sequence- or picture-level in existing coding techniques. When providing and communicating transform coefficient scaling at the block level, scaling matrices that adapt to block contents can be used to improve the visual acuity of a given block when encoding a video picture or still image, instead of having to select a scaling matrix that would be applied to the entire picture. This approach allows more detail to be preserved in video and image coding. |
