WO2021135983A1

WO2021135983A1 - Video transcoding method and apparatus, server and storage medium

Info

Publication number: WO2021135983A1
Application number: PCT/CN2020/137513
Authority: WO
Inventors: 刘晓娟
Original assignee: 百果园技术(新加坡)有限公司; 刘晓娟
Priority date: 2019-12-31
Filing date: 2020-12-18
Publication date: 2021-07-08
Also published as: CN111107395B; CN111107395A

Abstract

Disclosed are a video transcoding method and apparatus, a server and a storage medium. The method comprises: according to a picture encoding feature of a video to be transcoded, determining a transcoding factor selected and used when said video reaches, after being transcoded at each different rate gear to be used for transcoding, a subjective quality index specified by each of the rate gears to be transcoded; and by using the different rate gears to be transcoded and corresponding transcoding factors, transcoding said video.

Description

Method, device, server and storage medium for video transcoding

This application claims the priority of a Chinese patent application filed with the Chinese Patent Office with an application number of 201911410012.0 on December 31, 2019. The entire content of this application is incorporated into this application by reference.

Technical field

The embodiments of the present application relate to the field of video processing, for example, to a method, device, server, and storage medium for video transcoding.

Background technique

In a video transcoding system, in order to adapt to the network bandwidth of different terminals at different times and the corresponding terminal processing capabilities, it is often necessary to follow different bit rate gears and the preset constant bit rate factor ( Constant rate factor, CRF) performs multi-rate transcoding on the received source video stream. The CRF can make the source video stream reach the corresponding video playback quality after transcoding at this bit rate gear, so that it can be subsequently transferred to different terminals Distribute video streams in adaptive bitrate gears.

At this time, when the source video stream is transcoded with the corresponding preset CRF under different bit rate gears, the video playback quality is mainly the objective playback quality of the video, and due to the contrast sensitivity and brightness nonlinearity of the human eye to the video , Frequency sensitivity and masking effect, users will have a big difference in the subjective playback quality of different content video streams after transcoding with the same bitrate. For example, users are more sensitive to the distortion of the video picture under slow motion. , And it is not easy to detect the distortion of the video picture under intense exercise. Therefore, for any video, the corresponding preset fixed CRF is used for multi-bitrate transcoding at different bitrates. It cannot be guaranteed that the different videos watched by the user can reach The same subjective quality.

First, perform perceptual analysis on the content of the source video stream to determine the video category to which the source video stream belongs, such as movies, sports, or animation, etc., and then use the CRF pairs that are set in advance for different bitrates under the video category. The source video stream is transcoded at multiple bit rates. At this time, by setting the corresponding CRF under different video categories, the corresponding CRF is selected according to different user bandwidths to deliver the video, but it cannot accurately measure the user's different content under the same video category. The subjective playback quality of the video stream.

Summary of the invention

The embodiments of the application provide a method, device, server and storage medium for video transcoding to ensure that the subjective quality of different videos to be transcoded is consistent after being transcoded at the same bit rate gear, and the rationality of bit rate allocation is improved. .

The embodiment of the present application provides a method for video transcoding. The method includes: determining each of the to-be-transcoded video to be transcoded in different to-be-transcoded rate gears according to the picture coding characteristics of the to-be-transcoded video The transcoding factor used when the subjective quality index specified by each of the to-be-transcoding rate gears is reached after being transcoded in the rate gear; the different to-be-transcoded rate gears and the corresponding transcoding factors are used to compare the to-be-transcoded rate gears and the corresponding transcoding factors. Code video for transcoding.

The embodiment of the application provides a video transcoding device, the device includes: a transcoding factor determination module configured to determine the video to be transcoded at different transcoding rate files according to the picture coding characteristics of the video to be transcoded The transcoding factor used when each of the to-be-transcoded rate gears is transcoded and reaches the subjective quality index specified by each of the to-be-transcoded rate gears; the video transcoding module is set to use different to-be-transcoded The code rate gear and the corresponding transcoding factor transcode the video to be transcoded.

An embodiment of the present application provides a server, which includes: one or more processors; a storage device configured to store one or more programs; when the one or more programs are used by the one or more processors When executed, the one or more processors are caused to implement the video transcoding method described in any embodiment of the present application.

An embodiment of the present application provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the method for video transcoding described in any embodiment of the present application is implemented.

Description of the drawings

FIG. 1A is a flowchart of a method for video transcoding provided in Embodiment 1 of this application;

FIG. 1B is a schematic diagram of the principle of the video transcoding process provided in Embodiment 1 of this application;

2A is a flowchart of a video transcoding method provided in Embodiment 2 of this application;

2B is a schematic diagram of the principle of the video transcoding process provided in the second embodiment of the application;

3A is a flowchart of a video transcoding method provided in Embodiment 3 of this application;

FIG. 3B is a schematic diagram of the principle of the video transcoding process provided in Embodiment 3 of this application;

FIG. 4 is a schematic structural diagram of a video transcoding device provided in Embodiment 4 of this application;

FIG. 5 is a schematic structural diagram of a server provided in Embodiment 5 of this application.

Detailed ways

The application will be described below with reference to the drawings and embodiments. It can be understood that the embodiments described here are only used to explain the application, but not to limit the application. In addition, it should be noted that, for ease of description, the drawings only show a part of the structure related to the present application instead of all of the structure.

Example one

FIG. 1A is a flowchart of a method for video transcoding provided in Embodiment 1 of this application. This embodiment can be applied to the case of performing multi-rate transcoding on any video. The video transcoding method provided in this embodiment can be executed by the video transcoding device provided in this embodiment of the application, and the device can be implemented in software and/or hardware, and is integrated in the server that executes the method. , The server can be a back-end server that stores different video data.

Referring to FIG. 1A, the method may include the following steps:

S110, according to the picture coding characteristics of the video to be transcoded, determine that the video to be transcoded reaches each of the transcoding rate gears after being transcoded in each of the different transcoding rate gears to be transcoded The transcoding factor selected when the subjective quality index is specified.

In order to avoid video playback freezes, when multiple users download the corresponding video from the server, they need to select the video at the bit rate that matches the current network bandwidth, and because the user’s network bandwidth changes in real time during the video download process, in order to avoid Video playback freezes after downloading videos under multiple network bandwidths. For any source video, the server will perform multi-rate transcoding on the source video according to the preset multiple bit-rate gears, so that subsequent delivery to users will be adaptive. The source video at the bitrate.

The video to be transcoded is a source video that needs to be transcoded at multiple rates under any content type uploaded to the server by other users. Picture coding feature refers to the basic parameters included in the source video at a specific bit rate gear that can evaluate the objective encoding quality of multiple video frames in the source video after encoding at the specific bit rate gear, for example, the encoded video The peak signal to noise ratio (PSNR) of the frame, coding rate, and coding quantization parameters, etc. At the same time, the picture coding feature is a feature set that can characterize the temporal and spatial complexity of the video picture. At this time, the picture coding feature corresponds to The specific bitrate can be the coding bitrate used when the user uploads the source video, or the specific bitrate used by the server for preliminary transcoding after receiving the source video. The to-be-transcoded rate gears are a variety of transcoding rates set in advance for the source video that can match the real-time changing network bandwidth. The subjective quality index refers to the subjective playback quality required for the user’s viewing experience when the video to be transcoded is transcoded at different transcoding rate gears and then played on the user terminal. Due to the integration of video quality evaluation methods (Video Multimethod Assessment) Fusion (VMAF) algorithm can better measure the relationship between the source video content and the subjective viewing experience of different users. Therefore, the subjective quality index in this embodiment can be represented by VMAF scores to measure the subjective subjectiveness of the source video after transcoding. Playback quality.

Since the bit rate gear determines the overall file size of the video to be transcoded in a unit time, the transcoding quality of videos with different content cannot be guaranteed, so the transcoding factor in this embodiment is used for different transcoding rate gears. Control the transcoding quality of videos to be transcoded with different content. As shown in FIG. 1B, in this embodiment, a variety of different transcoding factors are preset under each rate gear to be transcoded, so as to subsequently filter out and specify the subjective quality under each gear rate to be transcoded. The transcoding factor adapted to the index is then used to transcode the source video under the to-be-transcoded rate gear. The transcoding factor in this embodiment may be a plurality of CRFs preset in each gear to be transcoded.

Optionally, after the server obtains the video to be transcoded, it first determines that the video to be transcoded is encoded at a specific bitrate, and can evaluate multiple pictures of the encoded video quality of the video to be transcoded. Encoding characteristics, and as shown in Figure 1B, in each to-be-transcoded rate gear, the to-be-transcoded rate gear and different transcoding factors set under the to-be-transcoded rate gear can be used for the to-be-transcoded rate gear. The transcoded video is transcoded multiple times with different transcoding factors under the same transcoding rate gear, and the picture encoding characteristics of the video to be transcoded are searched for after transcoding under different transcoding factors in the transcoding rate gear Corresponding feature transcoding effect, and determine whether the feature transcoding effect after selecting different transcoding factors in the to-be-transcoding rate gear reaches the subjective quality index specified by the to-be-transcoding rate gear. The transcoding factor selected when the specified subjective quality index is reached under the transcoding rate gear is used as the transcoding factor for the video to be transcoded in this embodiment to be adapted under the transcoding rate gear to be subsequently transcoded The video is actually transcoded; according to the above steps, the transcoding factor that the video to be transcoded adapts to under different transcoding rate gears is determined to ensure the accuracy of the selected transcoding factor under different bit rate gears.

S120, using different to-be-transcoded rate gears and corresponding transcoding factors to transcode the video to be transcoded.

Optionally, after determining the transcoding factor that the video to be transcoded is adapted to in each of the different transcoding rate gears to be transcoded, it can be used in each transcoding rate gear to be transcoded. The to-be-transcoded rate gear and the corresponding adapted transcoding factor are used to transcode the to-be-transcoded video to realize the multi-rate transcoding of the to-be-transcoded video, so as to allocate the bit rate reasonably and ensure that the different videos to be transcoded are in The subjective quality after transcoding in each of the different bit rate gears is consistent, avoiding unnecessary bit rate waste and saving bandwidth resources.

In the technical solution provided in this embodiment, the subjective quality index required to be achieved by the transcoded video is specified in advance under each to-be-transcoded rate gear, so that the same subjective quality is set for the video to be transcoded under the same bit-rate gear. Indicators to ensure that the subjective quality of the video to be transcoded remains the same after being transcoded at the same bit rate gear. At this time, in each gear to be transcoded, a variety of transcoding factors are selected for the transcoding of the video to be transcoded. Code, and determine the image encoding feature of the video to be transcoded, after transcoding with the cooperation of the corresponding transcoding factor, the selected transcoding factor to be used when the subjective quality index specified under the to-be-transcoding rate gear is reached, and then determine each A transcoding factor under the transcoding rate gear. At this time, the effect of any video to be transcoded after transcoding according to the preset transcoding factor is compared with the specified subjective quality index to accurately measure any The subjective playback quality of transcoded videos under different bit rate gears, to ensure the accuracy of the selected transcoding factors under different bit rate gears, and then use different to-be-transcoded rate gears and corresponding transcoding factors to treat transcoding The video is transcoded with multiple bit rates, thereby realizing the reasonable allocation of bit rates, avoiding unnecessary bit rate waste and saving bandwidth resources.

Example two

FIG. 2A is a flowchart of a video transcoding method provided in Embodiment 2 of this application, and FIG. 2B is a schematic diagram of the principle of a video transcoding process provided in Embodiment 2 of this application. This embodiment is described on the basis of the above-mentioned embodiment. This embodiment mainly explains the process of determining the transcoding factor adapted to the video to be transcoded under different to-be-transcoding rate gears.

Optionally, as shown in FIG. 2A, this embodiment may include the following steps:

S210: For each rate gear to be transcoded, the image encoding feature of the video to be transcoded, the rate gear to be transcoded, and the subjective quality index specified by the rate gear to be transcoded are merged to obtain the video to be transcoded in the The comprehensive transcoding feature under the rate to be transcoded.

Optionally, since this embodiment uses a pre-trained neural network model to determine the transcoding factors adapted under different transcoding rate gears, it is first necessary to obtain the video to be transcoded in multiple transcoding rate gears. For each rate gear to be transcoded, this embodiment can encode the image encoding feature of the video to be transcoded at a specific bit rate gear, and the bit rate value of the bit rate gear to be transcoded. And the subjective quality indicators specified in advance under the to-be-transcoded rate gear are feature fusion to generate the comprehensive transcoding features of the to-be-transcoded video under the to-be-transcoded rate gear; according to the above steps, the to-be-transcoded video is generated at each A comprehensive transcoding feature under the to-be-transcoding rate gear.

Exemplarily, as shown in FIG. 2B, in this embodiment, the picture encoding feature of the video to be transcoded, the to-be-transcoded rate gear, and the subjective quality index specified by the to-be-transcoded rate gear are combined to obtain the video to be transcoded The comprehensive transcoding feature under the to-be-transcoded rate gear may include: the picture encoding characteristics of the video to be transcoded, the to-be-transcoded rate gear, and the subjective quality index specified by the to-be-transcoded rate gear to expand dimensions Fusion, to obtain the comprehensive transcoding feature of the video to be transcoded at the rate to be transcoded.

In order to ensure the prediction accuracy of the neural network model, it is required that the comprehensive transcoding features of the video to be transcoded under different transcoding rates can include features in multiple dimensions, so that the subsequent fusion analysis of a large number of dimensional features; therefore In this embodiment, the image encoding feature of the video to be transcoded, the bit rate value of the to-be-transcoded rate gear, and the subjective quality index specified under the to-be-transcoded rate gear can be sequentially performed four-order cross product operation, and the to-be-transcoded video The picture coding features of the coded video, the to-be-transcoded rate gear, and the subjective quality index specified by the to-be-transcoded rate gear are merged, so that the characteristic dimension of the integrated transcoding characteristic after fusion is further compared with the directly merged characteristic dimension Expand to obtain a comprehensive transcoding feature of the video to be transcoded after dimension expansion and fusion at the to-be-transcoding rate gear. At this time, the comprehensive transcoding feature after dimension expansion and fusion contains a large amount of feature information in different dimensions.

S220: Perform a transcoding judgment on the comprehensive transcoding features of the video to be transcoded at different levels of the transcoding rate through the pre-built transcoding classification model, and determine that the video to be transcoded adapts under the different levels of the transcoding rate to be transcoded The transcoding factor.

Optionally, this embodiment selects a large number of video samples in advance, and sets the transcoding factor labels adapted to the multiple video samples under different to-be-transcoded rate gears, and then trains the transcoding classification model in this embodiment. Until the transcoding classification model can accurately determine the transcoding factor that any video adapts under different to-be-transcoding rate gears. After determining the comprehensive transcoding features of the video to be transcoded under different transcoding rate gears, as shown in Figure 2B, the comprehensive transcoding features under the different transcoding rate gears can be input into the pre-built transcoding in turn. In the code classification model, the transcoding classification model is used to make corresponding transcoding judgments on the comprehensive transcoding features under different transcoding rate gears, so as to determine the adaptation of the video to be transcoded under different transcoding rate gears. Transcoding factor.

When training the transcoding classification model in this embodiment, a large number of video samples will be selected first, and the subjective quality index of each video sample at different levels to be transcoded will be determined, and the subjective quality index will be selected when the subjective quality index is reached. The transcoding factor of, the transcoding factor is used as the corresponding sample label, and for each video sample, the video sample’s picture coding feature, the transcoding rate gear to be transcoded and the specified subjective quality index are subjected to a four-order cross product operation , Get the comprehensive transcoding feature after the expansion and fusion, and then use the comprehensive transcoding feature of a large number of video samples after the expansion and fusion as the corresponding training sample set, and multiply the training sample set through the initially set transcoding classification model. Classification training, and then continuously update the network parameters in the transcoding classification model until the training is completed.

Exemplarily, as shown in FIG. 2B, in this embodiment, the pre-built transcoding classification model is used to determine the comprehensive transcoding features of the video to be transcoded at different transcoding rate gears to determine the transcoding to be transcoded. The transcoding factor that the video adapts to different transcoding rate gears may include: for each transcoding rate gear, the comprehensive transcoding feature of the video to be transcoded in the transcoding rate gear is input In the transcoding classification model, the classification scores of the video to be transcoded under different preset transcoding factors are obtained; the preset transcoding factor corresponding to the video to be transcoded with the highest classification score is used as the video to be transcoded in the to-be-transcoded video The transcoding factor adapted in the rate gear.

In the transcoding classification model of this embodiment, multiple transcoding factors are preset, and the comprehensive transcoding features of the video to be transcoded at each transcoding rate gear to be transcoded are input into the transcoding classification model, and the transcoding The code classification model analyzes the comprehensive transcoding features of each rate to be transcoded to output the classification scores of the video to be transcoded under different preset transcoding factors in the transcoding classification model, and then select different The preset transcoding factor corresponding to the to-be-transcoded video with the highest classification score under the to-be-transcoding rate gear is used as the transcoding factor for the to-be-transcoded video to be adapted under the to-be-transcoding rate gear.

In order to reduce the amount of extra calculation in the transcoding classification model, the transcoding classification model in this embodiment may be composed of two or more layers of transcoding classification sub-models. For example, a three-layer small neural network is used. The first layer can be the series connection of full connection and convolution. The second and third layers are the series connection of batch normalization, full connection and convolution respectively. Finally, the logistic regression layer outputs the waiting The classification score of the transcoded video under each preset transcoding factor.

S230: Transcoding the video to be transcoded by using different to-be-transcoded rate gears and corresponding transcoding factors.

The technical solution provided in this embodiment integrates the picture coding characteristics of the video to be transcoded, the rate gear to be transcoded, and the designated subjective quality index to obtain the comprehensive transcoding characteristics of the video to be transcoded at different transcoding rate gears. , And use the pre-built transcoding classification model to make transcoding judgments on the comprehensive transcoding features under different transcoding rate gears, and determine the transcoding factor for the video to be transcoded under different transcoding rate gears. , To ensure the adaptation accuracy of the selected transcoding factors under different to-be-transcoded rate gears, thereby realizing the reasonable allocation of bit-rates, and ensuring that different to-be-transcoded videos are in the same bit-rate gear of different bit-rate gears The subjective quality after transcoding is consistent, avoiding unnecessary bit rate waste and saving bandwidth resources.

Example three

FIG. 3A is a flowchart of a method for video transcoding provided in Embodiment 3 of this application, and FIG. 3B is a schematic diagram of the principle of a video transcoding process provided in Embodiment 3 of this application. This embodiment is described on the basis of the foregoing embodiment. This embodiment mainly explains the extraction process of the picture coding features and subjective quality indicators of the video to be transcoded.

Optionally, as shown in FIG. 3A, this embodiment may include the following steps:

S310: Extract the corresponding picture encoding features of the video to be transcoded after being transcoded in the lowest bit rate gear among the different transcoding rate gears, and subjective quality indicators specified by the different transcoding rate gears.

Optionally, in this embodiment, before acquiring the image encoding feature of the video to be transcoded, the video to be transcoded is first transcoded at the lowest bit rate among the different to-be-transcoded rate gears, and the video is transcoded at the lowest bit rate. The basic characteristics of the picture are extracted from the transcoded video at the rate level, as the picture coding characteristics of the video to be transcoded in this embodiment, and the video content contained in the transcoded video is analyzed by the VMAF algorithm to determine the transcoded video The subjective quality index specified for the video at different levels of the transcoding rate to be used for subsequent judgment of the adapted transcoding factor.

Exemplarily, as shown in FIG. 3B, in this embodiment, extracting the corresponding picture coding features of the video to be transcoded at the lowest bit rate among the different to-be-transcoded rate gears may include: using different to-be-transcoded Transcode the video to be transcoded with the lowest bit rate gear in the bit rate gear and the fixed transcoding factor under the lowest bit rate gear; extract multiple keys of the video to be transcoded after being transcoded at the lowest bit rate gear The key information in the frame, the resolution and target quality of the video at different bit rate levels and the above-mentioned multiple key information, after a certain dimension expansion method, obtain the picture coding characteristics of the video to be transcoded.

First, select the lowest bit rate gear from the different to-be-transcoded rate gears. In this embodiment, the corresponding fixed transcoding factor is set for the lowest bit rate gear in advance. At this time, the lowest bit rate gear and the corresponding fixed transcoding factor are used. The transcoding factor transcodes the video to be transcoded, and extracts the key information that can characterize the picture characteristics from the transcoded video at the lowest bit rate gear, and compares the resolution and target quality of the video at different gears with the above key Information, through a certain dimension expansion method, the picture coding characteristics of the video to be transcoded are obtained.

S320: Determine, according to the image encoding characteristics of the video to be transcoded, that the video to be transcoded reaches each of the transcoding rate gears after being transcoded in each of the different transcoding rate gears. The transcoding factor selected in the case of the specified subjective quality index.

S330: Transcoding the video to be transcoded by using different to-be-transcoded rate gears and corresponding transcoding factors.

The technical solution provided in this embodiment obtains the corresponding picture coding features of the video to be transcoded at the lowest bit rate among the different to-be-transcoded rate gears, ensures the accuracy of the picture coding characteristics, and then determines the video to be transcoded. The picture encoding feature of the coded video is transcoded with the cooperation of the corresponding transcoding factor under the to-be-transcoded rate gear and the transcoding factor selected when the subjective quality index specified under the to-be-transcoded rate gear is reached. The effect of any video to be transcoded after being transcoded according to the predicted transcoding factor is compared with the specified subjective quality index to accurately measure the subjective playback quality of any video to be transcoded under different bitrates to ensure the difference The accuracy of the selected transcoding factor under the bit rate gear, and then use different to-be-transcoded rate gears and the corresponding transcoding factor to perform multi-rate transcoding on the video to be transcoded, thus realizing the reasonable allocation of the bit rate. Ensure that different videos to be transcoded have the same subjective quality after being transcoded under the same bitrate gears of different bitrate gears, avoid unnecessary bitrate waste and save bandwidth resources.

Example four

FIG. 4 is a schematic structural diagram of a video transcoding apparatus provided in Embodiment 4 of this application. As shown in FIG. 4, the device may include: a transcoding factor determining module 410, configured to determine each of the to-be-transcoded video in different to-be-transcoded rate gears according to the picture coding characteristics of the to-be-transcoded video The transcoding factor used when the subjective quality index specified by each of the transcoding rate gears is reached after being transcoded in the rate gear; the video transcoding module 420 is set to adopt different transcoding rate gears and The corresponding transcoding factor transcodes the video to be transcoded.

In the technical solution provided in this embodiment, the subjective quality index required to be achieved by the transcoded video is specified in advance under each to-be-transcoded rate gear, so that the same subjective quality is set for the video to be transcoded under the same bit-rate gear. Indicators to ensure that the subjective quality of the video to be transcoded remains the same after being transcoded at the same bit rate gear. At this time, in each gear to be transcoded, a variety of transcoding factors are selected for the transcoding of the video to be transcoded. Code, and determine the image encoding feature of the video to be transcoded, after transcoding with the cooperation of the corresponding transcoding factor, the selected transcoding factor to be used when the subjective quality index specified under the to-be-transcoding rate gear is reached, and then determine each A transcoding factor under the transcoding rate gear. At this time, the effect of any video to be transcoded after transcoding according to the preset transcoding factor is compared with the specified subjective quality index to accurately measure any The subjective playback quality of transcoded videos under different bit rate gears, to ensure the accuracy of the selected transcoding factors under different bit rate gears, and then use different to-be-transcoded rate gears and corresponding transcoding factors to treat transcoding The video is transcoded at multiple bit rates, thus realizing the reasonable allocation of bit rates, ensuring that the subjective quality of different videos to be transcoded under the same bit rate gear at different bit rate gears is consistent, and unnecessary codes are avoided. Rate waste and save bandwidth resources.

The above-mentioned transcoding factor determination module 410 may include: a feature fusion unit configured to fuse the picture coding features of the video to be transcoded, the to-be-transcoded rate gear, and the to-be-transcoded rate for each rate gear to be transcoded The subjective quality index specified by the gear position is used to obtain the comprehensive transcoding characteristics of the video to be transcoded under the transcoding rate gear position; the transcoding factor adaptation unit is set to treat the transcoded video separately through the pre-built transcoding classification model The comprehensive transcoding features under different transcoding rate gears are used for transcoding determination, and the transcoding factor adapted to the video to be transcoded under the different transcoding rate gears is determined.

The above-mentioned feature fusion unit is set to expand the image encoding feature of the video to be transcoded, the to-be-transcoded rate gear, and the subjective quality index specified by the to-be-transcoded rate gear for each of the to-be-transcoded rate gears. Dimension fusion is obtained to obtain the comprehensive transcoding feature of the video to be transcoded at the to-be-transcoded rate gear.

The above-mentioned transcoding factor adaptation unit is set to: for each rate gear to be transcoded, the comprehensive transcoding feature of the video to be transcoded in the rate gear to be transcoded is input into the transcoding classification model to obtain the The classification score of the transcoded video under different preset transcoding factors; the preset transcoding factor with the highest classification score is used as the transcoding factor for the video to be transcoded to be adapted under the to-be-transcoded rate gear.

The above-mentioned transcoding classification model may be composed of two or more layers of transcoding classification sub-models.

The above-mentioned video transcoding device may further include: a transcoding parameter extraction module, which is configured to extract the corresponding picture encoding features of the video to be transcoded at the lowest bit rate among different transcoding rate gears, and different The subjective quality index designated by the transcoding rate gear.

The above-mentioned transcoding parameter extraction module can be set to extract the corresponding picture encoding characteristics of the video to be transcoded at the lowest bit rate position among the different to be transcoded rate gears through the following method: use different to be transcoded rate files Transcode the video to be transcoded in the lowest bit rate gear and the fixed transcoding factor under the lowest bit rate gear; extract multiple key frames of the video to be transcoded after being transcoded in the lowest bit rate gear, and The key information in the multiple key frames is combined to obtain the picture coding feature of the video to be transcoded, and the key information is used to characterize the picture coding feature of the video to be transcoded.

The video transcoding device provided in this embodiment is applicable to the video transcoding method provided in any of the foregoing embodiments, and has corresponding functions.

Example five

FIG. 5 is a schematic structural diagram of a server provided in Embodiment 5 of this application. As shown in FIG. 5, the server includes a processor 50, a storage device 51, and a communication device 52; the number of processors 50 in the server may be one or more. One, one processor 50 is taken as an example in FIG. 5; the processor 50, the storage device 51, and the communication device 52 in the server may be connected by a bus or other means. In FIG. 5, the connection by a bus is taken as an example.

As a computer-readable storage medium, the storage device 51 can be configured to store software programs, computer-executable programs, and modules, such as program instructions/modules corresponding to the video transcoding method provided in the embodiments of the present application. The processor 50 executes various functional applications and data processing of the server by running the software programs, instructions, and modules stored in the storage device 51, that is, realizes the above-mentioned video transcoding method.

The storage device 51 may mainly include a storage program area and a storage data area. The storage program area may store an operating system and an application program required by at least one function; the storage data area may store data created according to the use of the terminal, and the like. In addition, the storage device 51 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage devices. In some examples, the storage device 51 may further include a memory remotely provided with respect to the processor 50, and these remote memories may be connected to the server through a network. Examples of the aforementioned networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

The communication device 62 may be configured to implement a network connection or a mobile data connection between the server and the terminal.

The server provided in this embodiment can be configured to execute the video transcoding method provided in any of the foregoing embodiments, and has corresponding functions.

Example Six

The sixth embodiment of the present application also provides a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the video transcoding method in any of the foregoing embodiments can be implemented. The method may include: determining, according to the picture coding characteristics of the video to be transcoded, that the video to be transcoded is transcoded in each of the different to-be-transcoded rate gears to reach each of the to-be-transcoded rate gears. Transcoding factor selected in the case of subjective quality indicators specified by the transcoding rate gear; using different transcoding rate gears and corresponding transcoding factors to transcode the video to be transcoded.

Of course, the storage medium containing computer-executable instructions provided by the embodiments of the present application is not limited to the method operations described above, and can also execute the video transcoding method provided by any embodiment of the present application. Related operations in.

From the above description of the implementation manners, those skilled in the art can understand that this application can be implemented by software and necessary general-purpose hardware, and of course, it can also be implemented by hardware. Based on this understanding, the technical solution of the present application can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), Random Access Memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., including multiple instructions to make a computer device (which can be a personal computer, server, or network device, etc.) execute any of this application The method described in the embodiment.

It is worth noting that, in the above embodiment of the video transcoding device, the multiple units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; In addition, the specific names of multiple functional units are only used to facilitate distinguishing from each other, and are not used to limit the protection scope of the present application.

Claims

A method for video transcoding, including:

According to the picture encoding characteristics of the video to be transcoded, it is determined that the video to be transcoded reaches each of the transcoding rate gears after being transcoded in each of the different transcoding rate gears. The transcoding factor selected in the case of the specified subjective quality index;

Transcoding the video to be transcoded using different to-be-transcoded rate gears and corresponding transcoding factors.
The method according to claim 1, wherein the video to be transcoded is determined to be transcoded at each of the different transcoding rate gears after transcoding according to the picture encoding characteristics of the video to be transcoded The transcoding factor used in the case of reaching the subjective quality index specified by each of the to-be-transcoding rate gears includes:

Combining the picture coding features of each video to be transcoded, each of the to-be-transcoded rate gears, and the subjective quality index specified by each of the to-be-transcoded rate gears, to obtain the video to be transcoded in each of the A comprehensive transcoding feature under the to-be-transcoding rate gear;

The comprehensive transcoding features of the video to be transcoded under different transcoding rate gears are respectively determined through the pre-built transcoding classification model, and it is determined that the video to be transcoded is in the different transcoding rate gears. The transcoding factor for the next adaptation.
3. The method according to claim 2, wherein the image coding characteristics of each video to be transcoded, each of the to-be-transcoded rate gears, and the subjective quality index designated by each of the to-be-transcoded rate gears are merged, Obtaining the comprehensive transcoding features of the video to be transcoded in each of the transcoding rate gears to be transcoded includes:

The image encoding feature of each video to be transcoded, each of the to-be-transcoded rate gears, and the subjective quality index specified by each of the to-be-transcoded rate gears are expanded and fused to obtain the video to be transcoded The comprehensive transcoding feature in each of the to-be-transcoding rate gears.
The method according to claim 2, wherein the comprehensive transcoding features of the to-be-transcoded video at different to-be-transcoding rate gears are respectively subjected to transcoding judgments through a pre-built transcoding classification model, and the to-be-transcoding rate is determined to be The transcoding factor adapted for the transcoded video under different levels of the transcoding rate to be transcoded, including:

Input the comprehensive transcoding features of the video to be transcoded under each transcoding rate gear into the transcoding classification model to obtain the classification scores of the video to be transcoded under different preset transcoding factors;

The preset transcoding factor with the highest classification score is used as the transcoding factor adapted to the video to be transcoded in each of the transcoding rate gears to be transcoded.
The method according to claim 2, wherein the transcoding classification model is composed of at least two layers of transcoding classification sub-models.
The method according to any one of claims 1 to 5, wherein after determining that the video to be transcoded is transcoded in each of the different to-be-transcoding-rate gears, it is reached after transcoding at each of the different-to-transcoding-rate gears. Before the subjective quality index to be used in the case of the subjective quality index specified by the transcoding rate gear, it also includes:

Extract the corresponding picture coding features of the video to be transcoded after being transcoded at the lowest bit rate gear among the different to-be-transcoded rate gears, and subjective quality indicators specified by the different to-be-transcoded rate gears.
The method according to claim 6, wherein extracting the corresponding picture coding features of the video to be transcoded after being transcoded at the lowest bit rate among the different to be transcoded rate gears, comprises:

Transcoding the video to be transcoded by using the lowest bit rate gear of the different to-be-transcoded rate gears and the fixed transcoding factor under the lowest bit rate gear;

Extracting multiple key frames of the video to be transcoded after being transcoded at the lowest bit rate gear, and merging the key information in the multiple key frames to obtain the picture coding characteristics of the video to be transcoded, The key information is used to characterize the picture coding characteristics of the video to be transcoded.
A video transcoding device includes:

The transcoding factor determination module is configured to determine that the video to be transcoded is transcoded at each of the different transcoding rate gears to reach the said transcoding factor according to the picture coding characteristics of the video to be transcoded. The transcoding factor used in the case of the subjective quality index specified by each transcoding rate gear;

The video transcoding module is configured to transcode the video to be transcoded by using different to-be-transcoded rate gears and corresponding transcoding factors.
A server that includes:

At least one processor;

The storage device is set to store at least one program;

When the at least one program is executed by the at least one processor, the at least one processor is caused to implement the video transcoding method according to any one of claims 1-7.
A computer-readable storage medium storing a computer program, which when executed by a processor, realizes the video transcoding method according to any one of claims 1-7.