视频编解码芯片设计原理----02 视频编码技术框架与标准
![](http://static.careerengine.us/api/aov2/http%3A_%7C__%7C_mmbiz.qpic.cn_%7C_mmbiz_jpg_%7C_Pu7P4cXKvJXw9THSP7uamoY6moic3HI0kHicFFgLNBEbjGxiaHupn8iaaVJ71VdIoRwrib7xmtCXaHMa0fc3wazhNicQ_%7C_0%3Fwx_fmt%3Djpeg.jpg)
![](http://static.careerengine.us/api/aov2/https%3A_%7C__%7C_mmbiz.qpic.cn_%7C_mmbiz_png_%7C_lYDNMgmFjur6mfkPywZNgKZYXia5C4lA8Kd6CUYOibay3HbIfxBnNXH3v6jQZf6QquSL5dt6qYMrtQGh7go67FZg_%7C_640%3Fwx_fmt%3Dpng.jpg)
本文将介绍视频编码技术的混合编码框架和国内外主流编码标准的发展历程与特点。
![](http://static.careerengine.us/api/aov2/https%3A_%7C__%7C_mmbiz.qpic.cn_%7C_mmbiz_png_%7C_PPBibXczyTGQLbQ7DDcF2D79CDScRMVVJYNfVfgrEMmoYPUvWmvwibSeWS6pWwzLL5QcRkiaMXnVrJZEFFLoQ3oHA_%7C_640%3Fwx_fmt%3Dpng.jpg)
1
视频编码技术框架
![](http://static.careerengine.us/api/aov2/https%3A_%7C__%7C_mmbiz.qpic.cn_%7C_mmbiz_png_%7C_Pu7P4cXKvJXw9THSP7uamoY6moic3HI0kLHQFK38EQEhiaVuiaUVY9euFdZ5y2FA8x8klPw43SSZ6Zl7kwCeR8m7g_%7C_640%3Fwx_fmt%3Dpng.jpg)
2
视频编码标准
![](http://static.careerengine.us/api/aov2/https%3A_%7C__%7C_mmbiz.qpic.cn_%7C_mmbiz_png_%7C_Pu7P4cXKvJXw9THSP7uamoY6moic3HI0kibap1qN39zQwM65fzMflcskYqib7bBtH92icx3jwB3IlYmq8lC9FPcGVA_%7C_640%3Fwx_fmt%3Dpng.jpg)
3
视频编码技术发展
块划分
CTU最大尺寸为128x128 CU划分除四叉树外,还支持二叉树和三叉树划分 亮度和色度CTB可以独立划分
帧内预测
- 常规预测模式:65个角度模式、28个宽角度模式
- 矩阵加权帧内预测 (MIP, Matrix-based intra prediction)
- 多参考线帧内预测 (MRL, Multi-reference lines intraprediction)
- 帧内子块划分 (ISP, Intra sub-partitions)
- 跨分支线性预测 (CCLM, Cross-component linear model)
- 位置相关的组合预测 (PDPC, Position-dependent prediction combination)
帧间预测
- 新的候选项种类:History based MV、Pairwise MV等
- 新的运动模型:仿射 (Affine)
- 新的MVD编码方式①自适应运动适量精度 (AMVR, Adaptive MV resolution)②带有运动矢量差值的Merge模式 (MMVD, Merge MVD)③对称MVD模式 (SMVD, Symmetric MVD)④解码端运动矢量修正 (DMVR, Decoder-side MV refinement)
- 新的预测模式双向光流(BDOF, Bi-directional optical flow)、光流预测细化 (PROF, Prediction refinement with opticalflow)
![](http://static.careerengine.us/api/aov2/https%3A_%7C__%7C_mmbiz.qpic.cn_%7C_mmbiz_png_%7C_Pu7P4cXKvJW2IrM42Nmic79BaLwUz80ZZcHmKb5btqvrqbU50T2Iyz4T6pgkEkdRMukbAStiaypbhXG3G9HPJFkg_%7C_640%3Fwx_fmt%3Dpng.jpg)
变换
- 多变换核选择 (MTS, Multiple type transform)
- 大尺寸变换块高频系数置零
- 低频不可分离变换 (LFNST, Low-frequency non-separable transform)
- 子块变换 (SBT, Sub-block trans-forms)
量化
- 依赖标量量化 (DQ, Dependent quantization)
重建环路与环路滤波
- 自适应环路滤波 (ALF, Adaptive loop filter)
- 跨分量ALF (CCALF, Cross-component adaptive loopfilter)
- 环路重整形 (LMCS, Luma mapping with chroma scaling)
![](http://static.careerengine.us/api/aov2/https%3A_%7C__%7C_mmbiz.qpic.cn_%7C_mmbiz_png_%7C_Pu7P4cXKvJXw9THSP7uamoY6moic3HI0ktnHdiaZUuarfwdItAGfxgQBsp5qB1OyG2QZ6CG4kNo7icVvJicHOYtxiag_%7C_640%3Fwx_fmt%3Dpng.jpg)
其他
- 色度残差联合编码 (JCCR, Joint coding of chroma residual)
- 参考图像重采样 (RPR, Reference picture resampling)
![](http://static.careerengine.us/api/aov2/https%3A_%7C__%7C_mmbiz.qpic.cn_%7C_mmbiz_png_%7C_PPBibXczyTGQLbQ7DDcF2D79CDScRMVVJYNfVfgrEMmoYPUvWmvwibSeWS6pWwzLL5QcRkiaMXnVrJZEFFLoQ3oHA_%7C_640%3Fwx_fmt%3Dpng.jpg)
![](http://static.careerengine.us/api/aov2/https%3A_%7C__%7C_mmbiz.qpic.cn_%7C_mmbiz_png_%7C_GAibT6GPsHibekPWrcvgkzCjkMf4CHESNqHkdFwwT8bBV6KS25bsLXzQd5WXFEoic4BHkAVfUmibSQG786YhloMvBw_%7C_640%3Fwx_fmt%3Dpng.jpg)
![](http://static.careerengine.us/api/aov2/https%3A_%7C__%7C_mmbiz.qpic.cn_%7C_mmbiz_png_%7C_ULKUSjXeQbxecm8xLaBwx8No4ULvOhia5wJqF9cCQnYk9Aibu8dEaFESZL4SDosssjOmzEV3iaLuRbO64mWoZlHrg_%7C_640%3Fwx_fmt%3Dpng.jpg)
![](http://static.careerengine.us/api/aov2/https%3A_%7C__%7C_mmbiz.qpic.cn_%7C_mmbiz_png_%7C_dspHZmfzKI7uUX3DolTQgUibo84Z0ZibMj3Na5RRlXqWdR39qsBicvWhJuWzwKic0eWlJlNs3VcEYatfSdANGFlSGA_%7C_640%3Fwx_fmt%3Dpng.jpg)
关注我们
![](http://static.careerengine.us/api/aov2/https%3A_%7C__%7C_mmbiz.qpic.cn_%7C_mmbiz_png_%7C_iaNLDl7z9qYuhFDibF1zPTzZK3iab78swI3icZib1mluz5GPlkuibMFNueaSsdDDicXOicKQn1pY7jB5sBicibicOrDJgf96A_%7C_640%3Fwx_fmt%3Dpng.jpg)
![](http://static.careerengine.us/api/aov2/https%3A_%7C__%7C_mmbiz.qpic.cn_%7C_mmbiz_png_%7C_BHWB04Dh79icN5RLVsqbjeRAficDQjlq2iaH9987VaCUnZM8fXTthLl3V32HWia3jhTfk6qaItRcZ2R7m6kzDrc7sw_%7C_640%3Fwx_fmt%3Dpng.jpg)
最新评论
推荐文章
作者最新文章
你可能感兴趣的文章
Copyright Disclaimer: The copyright of contents (including texts, images, videos and audios) posted above belong to the User who shared or the third-party website which the User shared from. If you found your copyright have been infringed, please send a DMCA takedown notice to [email protected]. For more detail of the source, please click on the button "Read Original Post" below. For other communications, please send to [email protected].
版权声明:以上内容为用户推荐收藏至CareerEngine平台,其内容(含文字、图片、视频、音频等)及知识版权均属用户或用户转发自的第三方网站,如涉嫌侵权,请通知[email protected]进行信息删除。如需查看信息来源,请点击“查看原文”。如需洽谈其它事宜,请联系[email protected]。
版权声明:以上内容为用户推荐收藏至CareerEngine平台,其内容(含文字、图片、视频、音频等)及知识版权均属用户或用户转发自的第三方网站,如涉嫌侵权,请通知[email protected]进行信息删除。如需查看信息来源,请点击“查看原文”。如需洽谈其它事宜,请联系[email protected]。