CN-117115511-B - Video processing method, device, electronic equipment and medium

Abstract

The invention relates to the technical field of computers, in particular to a video processing method, a device, electronic equipment and a medium, which are used for grouping videos contained in a video set to be processed to obtain a plurality of groups of video pairs, calibrating a first video and a second video contained in the video pair through a plurality of preset calibration models to obtain a plurality of calibration information for any video pair, wherein the calibration information comprises first calibration information of the first video and second calibration information of the second video, determining an uncertainty rate of the video pair based on the first calibration information and the second calibration information in the calibration information, wherein the uncertainty rate is used for representing difficulty of calibrating the video pair, and determining the video pair with the uncertainty rate meeting calibration conditions as a target video pair so as to obtain a target video set based on the target video pair. And screening the video pairs through the uncertainty, so as to obtain target video pairs meeting the calibration conditions, and improving the effectiveness of the target video set.

Inventors

• Zheng Chuanchuan
• SUN MING

Assignees

• 北京达佳互联信息技术有限公司

Dates

Publication Date

20260512

Application Date

20230721

Claims (9)

1. 1. A method of video processing, the method comprising: grouping videos contained in the video set to be processed to obtain a plurality of groups of video pairs; For any video pair, calibrating a first video and a second video contained in the video pair respectively through a plurality of preset calibration models to obtain a plurality of calibration information, wherein the calibration information comprises first calibration information of the first video and second calibration information of the second video; determining an uncertainty rate of the video pair based on the first calibration information and the second calibration information in the calibration information, wherein the uncertainty rate is used for representing the difficulty of calibrating the video pair; Determining a video pair with the uncertainty meeting the calibration condition as a target video pair so as to obtain a target video set based on the target video pair; The determining the uncertainty rate of the video pair based on the first calibration information and the second calibration information in each calibration information includes: Under the condition that the absolute value of the information value of each calibration information is larger than a specified threshold value, taking the calibration information comprising the first calibration information which is larger than the second calibration information as a first calibration result, and taking the calibration information comprising the first calibration information which is not larger than the second calibration information as a second calibration result, wherein the information value of the calibration information is the difference value between the first calibration information and the second calibration information in the calibration information; determining the number of the first calibration results as a first number and the number of the second calibration results as a second number; An uncertainty rate of the video pair is determined based on the first number, the second number, and a total number of the first number and the second number.
2. 2. The method of claim 1, wherein the calibration condition is that the uncertainty ratio is equal to 1/2.
3. 3. The method according to claim 1, wherein grouping the videos included in the set of videos to be processed to obtain multiple sets of video pairs includes: Identifying video attributes of each video contained in the video set to be processed; and dividing two videos with the same video attribute into the same group based on the video attribute of each video to obtain the multiple groups of video pairs.
4. 4. The method according to claim 1, wherein the calibrating the first video and the second video contained in the video pair by the plurality of preset calibration models to obtain a plurality of calibration information includes: Respectively inputting the first video and the second video into the plurality of preset calibration models so that the plurality of preset calibration models respectively calibrate the first video and the second video; and acquiring output data of each of the plurality of preset calibration models as the plurality of calibration information.
5. 5. The method of claim 1, wherein the plurality of pre-set calibration models are generated based on a plurality of subsets of data, the method further comprising: respectively sampling the appointed video set according to different sampling strategies; and for any sampling strategy, determining the video subset obtained by sampling according to the sampling strategy as a data subset, and obtaining the plurality of data subsets.
6. 6. The method according to claim 5, characterized in that the sampling positions and/or the sampling numbers of the different sampling strategies are different.
7. 7. A video processing apparatus, the apparatus comprising: the grouping module is configured to perform grouping on videos contained in the video set to be processed to obtain a plurality of groups of video pairs; The video processing device comprises a video pair, a calibration module, a video processing module and a video processing module, wherein the video pair comprises a first video and a second video, and the video pair comprises a plurality of video images, a plurality of video images and a plurality of video images, and the video images are respectively displayed in the video image pair; An uncertainty determination module configured to perform determining an uncertainty of the video pair based on the first calibration information and the second calibration information in each of the calibration information, the uncertainty being used to characterize a difficulty of calibrating the video pair; a target determining module configured to perform determining a video pair whose uncertainty rate satisfies a calibration condition as a target video pair to obtain a target video set based on the target video pair; Wherein the uncertainty determination module is specifically configured to perform: Under the condition that the absolute value of the information value of each calibration information is larger than a specified threshold value, taking the calibration information comprising the first calibration information which is larger than the second calibration information as a first calibration result, and taking the calibration information comprising the first calibration information which is not larger than the second calibration information as a second calibration result, wherein the information value of the calibration information is the difference value between the first calibration information and the second calibration information in the calibration information; determining the number of the first calibration results as a first number and the number of the second calibration results as a second number; An uncertainty rate of the video pair is determined based on the first number, the second number, and a total number of the first number and the second number.
8. 8. An electronic device, comprising: A processor; a memory for storing the processor-executable instructions; Wherein the processor is configured to execute the instructions to implement the method of any one of claims 1 to 6.
9. 9. A computer readable storage medium, characterized in that instructions in the storage medium, when executed by a processor of an electronic device, cause the electronic device to perform the method of any of claims 1 to 6.

Description

Video processing method, device, electronic equipment and medium Technical Field The disclosure relates to the field of computer technology, and in particular, to a video processing method, a video processing device, electronic equipment and a medium. Background With the development of computer technology, various types of data generated on a network, including video data, are increasing. For some video platforms, the video on the platform is usually required to be calibrated, so that the video is further processed according to the calibrated score, and accordingly, after a calibration model is trained by using a training video, the video to be calibrated is calibrated by directly using the calibration model after the training is completed. However, in the prior art, the video pairs in the training video are often randomly constructed, and a large number of video pairs which are easy to calibrate exist, which results in poor training effect when using such training video, so how to improve the effectiveness of the training video is a problem to be solved. Disclosure of Invention The disclosure provides a video processing method, a video processing device, electronic equipment and a medium, so as to at least solve the problem of how to improve the effectiveness of training videos. The technical scheme of the present disclosure is as follows: According to a first aspect of an embodiment of the present disclosure, there is provided a video processing method, including: grouping videos contained in the video set to be processed to obtain a plurality of groups of video pairs; For any video pair, calibrating a first video and a second video contained in the video pair respectively through a plurality of preset calibration models to obtain a plurality of calibration information, wherein the calibration information comprises first calibration information of the first video and second calibration information of the second video; determining an uncertainty rate of the video pair based on the first calibration information and the second calibration information in the calibration information, wherein the uncertainty rate is used for representing the difficulty of calibrating the video pair; and determining the video pair with the uncertainty meeting the calibration condition as a target video pair so as to obtain a target video set based on the target video pair. Optionally, the determining the uncertainty rate of the video pair based on the first calibration information and the second calibration information in each calibration information includes: Under the condition that the absolute value of the information value of each calibration information is larger than a specified threshold value, taking the calibration information comprising the first calibration information which is larger than the second calibration information as a first calibration result, and taking the calibration information comprising the first calibration information which is not larger than the second calibration information as a second calibration result, wherein the information value of the calibration information is the difference value between the first calibration information and the second calibration information in the calibration information; determining the number of the first calibration results as a first number and the number of the second calibration results as a second number; An uncertainty rate of the video pair is determined based on the first number, the second number, and a total number of the first number and the second number. Optionally, the calibration condition is that the uncertainty rate is equal to 1/2. Optionally, the grouping the videos included in the video set to be processed to obtain multiple groups of video pairs includes: Identifying video attributes of each video contained in the video set to be processed; and dividing two videos with the same video attribute into the same group based on the video attribute of each video to obtain the multiple groups of video pairs. Optionally, the calibrating the first video and the second video contained in the video pair through a plurality of preset calibration models to obtain a plurality of calibration information includes: Respectively inputting the first video and the second video into the plurality of preset calibration models so that the plurality of preset calibration models respectively calibrate the first video and the second video; and acquiring output data of each of the plurality of preset calibration models as the plurality of calibration information. Optionally, the plurality of preset calibration models are generated based on a plurality of data subsets, and the method further comprises: respectively sampling the appointed video set according to different sampling strategies; and for any sampling strategy, determining the video subset obtained by sampling according to the sampling strategy as a data subset, and obtaining the plurality of data subsets. Optiona