Shulou(Shulou.com)05/31 Report--

This article shows you how H.265 reduces the cost of video storage and transmission. The content is concise and easy to understand, and it will definitely make your eyes shine. I hope you can get something through the detailed introduction of this article.

In the Internet era, every technological innovation will bring commercial value and promote social progress. However, the strong market demand will also reverse the maturity of the technology.

From the perspective of the video industry, the richness of content is increasing. H.264 once made 1080p storm sweep the world, but with the popularity of 4K, VR and other industries, it can no longer lead the trend. Accordingly, in the process of video experience, in order to ensure clarity, the requirement of video bit rate is also increasing. The same bandwidth should have a higher bit rate, and the appearance of picture quality and stutter is bound to affect the user experience. On the contrary, if you want to develop, the cost of bandwidth is a problem that the industry can't get around. As a new generation of video coding standard, H.265 conforms to the times.

Today, let's take a look at how H.265 is used in LVB scenarios:

H.265 (HEVC) is a new generation of video coding algorithm. Compared with H.264, it can save the cost of video storage and transmission at the same definition. At the same time, the reduction of network bandwidth can also bring users a better video viewing experience under the same bandwidth. Let's take a look at how H.265 technology is implemented in live streaming. The following figure shows a typical LVB process:

L Green connection indicates the protocols supported by H.265

L Red connection indicates a protocol that is not supported by H.265

L Custom indicates the internal transmission protocol of the video live streaming system, which you can choose freely.

When it comes to video coding, it is possible to use H.265. In the above process, we see that there are two places related to coding, namely, providing H.265 coding capability on the encoder side and keeping H.264 on the codec side, and transcoding capability from H.264 to H.265 on the cloud. In order to use these two schemes, the cloud must do:

L supports H.265 in transmission. Currently, almost all LVB uplinks use RTMP, while downlinks usually use RTMP,HTTPFLV or HLS. However, the video encapsulation used by RTMP,HTTPFLV is that FLV,FLV does not support H.265, so FLV needs to be extended. TS supports H.265 and does not require extension.

Real-time transcoding from H.264 to H.265 is supported.

Let's take a look at how to support the encapsulation structure of H.265:FLV in FLV as tag one by one, and each tag has a tagheader indicating whether the tag is video, audio, or script. For video tag, a 4bit field identifies the video encoding type, as shown in the following figure:

As you can see, six types have been used for the field CodecID of this 4bit (0-15), and the rest can be used for extension. At the same time, if we look at ffmpeg's definition of FLV format, we will find that ffmpeg has used FLV CodecID up to 9, as shown in the following figure:

In order to avoid the compatibility problems caused by ffmpeg's extension of FLV, it is necessary to use other CodecID to represent H.265.

After using the extended transport protocol and implementing the transcoding of H265 in the cloud, the LVB process is as follows:

You can use the following combinations according to the capabilities of different push end or player end:

Push H.265 encoding, and use the extended RTMP protocol to upload H.265 live streams, convert the cloud to H.265 and H.264 with different bit rates, and output HLS, standard RTMP, extended RTMP, standard HTTPFLV, extended HTTPFLV for terminals with different decoding capabilities and network bandwidths.

The push uses H.264 encoding, and the cloud transcodes H.264 into H.264 and H.265 bitstreams with different bit rates, and outputs HLS, standard RTMP, extended RTMP, standard HTTPFLV, and extended HTTPFLV for terminals with different decoding capabilities and network bandwidths.

Admittedly, the birth of the new standard, accompanied by the applause and pursuit of flowers, also has problems. The complexity of the algorithm that H.265 depends on determines that it needs more computing power to support, followed by an increase in hardware consumption.

It is understood that the theoretical computation of H.265 decoding is 1.7 times that of H.264. Previously, dual-core A9 + single-core GPU configuration can play H.264 video smoothly, but when it comes to H.265 video, you can only enjoy a slide show. This is the main reason why all video cloud manufacturers are still in the exploratory stage.

Even so, H.265 can save 20% of the bitrate by 40% compared with H.264. Therefore, the purpose of applying H.265 in LVB is very clear, that is, to reduce video bitrate and save bandwidth costs.

Cloud computing technology is becoming more and more mature, and more and more manufacturers will pour into the video cloud market in the future. With the gathering of Internet giants such as BAT, the continuous application of technologies and standards such as big data, H.265, 4K, VR and AR also put forward higher requirements for video cloud services. How to adapt to technological innovation and help customers reduce costs according to their own advantages is a proposition worth thinking about. I believe that each manufacturer has its own answer.

The above is how H.265 reduces the cost of video storage and transmission. Have you learned any knowledge or skills? If you want to learn more skills or enrich your knowledge reserve, you are welcome to follow the industry information channel.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.