Automatic adjustment method, device and equipment for beautification in live broadcast of anchor and medium

Abstract

The invention provides an anchor live broadcast beautification automatic adjustment method and device, equipment and a medium, and the method comprises the steps: obtaining an initial beautification setting based on a preset beautification parameter; processing the live video stream based on the initial beauty setting to obtain an initial video stream; pushing the initial video stream to a third-party platform; monitoring the change of a subsequent video stream in real time; when it is monitored that the change of the subsequent video stream reaches a preset condition, obtaining a current beautifying parameter; based on the current facial beautification parameters, adjusting system facial beautification settings to obtain current facial beautification settings; processing a subsequent video stream based on the current facial beautification setting to obtain a current video stream; pushing the current video stream to a third-party platform; and repeating the step from real-time monitoring to pushing the current video stream to the third-party platform until the live broadcast is finished. The scheme can automatically adapt to the beauty requirements of different anchors, and dynamically adjust the beauty parameters according to the live broadcast real-time state, so as to improve the live broadcast management efficiency, and enhance the visual effect and audience experience.

Claims

1. A method for automatically adjusting the host's facial beauty during live broadcast, characterized in that: include: Based on the preset beauty parameters, adjust the system beauty settings to obtain the initial beauty settings; Based on the initial beauty setting, the live video stream is processed to obtain an initial video stream; Pushing the initial video stream to a third-party platform; Monitor the changes of subsequent video streams in real time; When it is detected that the change of the subsequent video stream reaches a preset condition, obtaining the current beauty parameter; Based on the current beauty parameters, adjust the system beauty settings to obtain the current beauty settings; Based on the current beauty setting, processing the subsequent video stream to obtain a current video stream; Pushing the current video stream to the third-party platform; Repeat the step of real-time monitoring of changes in subsequent video streams, to the step of pushing the current video stream to the third-party platform, until the live broadcast ends.

2. The method for automatically adjusting the host's facial beauty during live broadcast according to claim 1, characterized in that: After the live video stream is processed based on the initial beauty setting to obtain the initial video stream, the method further includes: The initial video stream is frame-cut to obtain an initial image; the initial image carries initial face information.

3. The method for automatically adjusting the host's facial beauty during live broadcast according to claim 2, characterized in that: The real-time monitoring of subsequent video stream changes includes: Obtaining the subsequent video stream; Extract frames from the subsequent video stream to obtain a current image; Analyze the difference between the current image and the initial image or the previous current image to obtain a monitoring result; wherein, before entering the loop, analyze the difference between the current image and the initial image, and after entering the loop, analyze the difference between the current image and the previous current image.

4. The method for automatically adjusting the host's facial beauty during live broadcast according to claim 3, characterized in that: When the subsequent video stream change is detected to meet a preset condition, obtaining the current beauty parameter includes: When the monitoring result indicates that the current image is different from the face in the initial image or the previous current image, face recognition is performed on the current image to obtain a current recognition result, so that the current image carries current face information; Based on the current recognition result, the preset beauty parameters matching the current recognition result are retrieved as the current beauty parameters.

5. The method for automatically adjusting the host's facial beauty during live broadcast according to claim 4, characterized in that: The acquiring of current beauty parameters when the subsequent video stream change is detected to meet a preset condition also includes: When the monitoring result indicates that the current image is different from the initial image or the previous current image in a state, it is determined whether there is an image whose error value with the current image is less than a preset threshold; wherein the limited range of the state is determined based on the beauty requirement; If not, then comparing and analyzing the changes in facial features and overall attributes of the person between the current image and the initial image or the last current image to obtain a comparison result; Based on the comparison result, the preset beauty parameters or the previous current beauty parameters are adjusted to obtain adjusted beauty parameters; wherein the preset beauty parameters are adjusted before entering the loop, and the previous current beauty parameters are adjusted after entering the loop; The adjusted beautification parameters are used as the current beautification parameters.

6. The method for automatically adjusting the host's facial beauty during live broadcast according to claim 5, characterized in that: The method further comprises: If so, retrieve the beauty parameters corresponding to the image whose error value with the current image is less than the preset threshold as the current beauty parameters.

7. A method for automatically adjusting the host's facial beauty during live broadcast according to any one of claims 1 to 6, characterized in that: Before adjusting the system beauty settings based on the preset beauty parameters to obtain the initial beauty settings, the method further includes: Obtaining the live video stream; Extract frames from the live video stream to obtain a live image; Performing face recognition on the live broadcast image to obtain an initial recognition result; Based on the initial recognition result, the preset beauty parameters matching the initial recognition result are retrieved.

8. A host live broadcast beauty automatic adjustment device, using the host live broadcast beauty automatic adjustment method according to any one of claims 1 to 7, characterized in that: The device comprises: An initial beauty setting module, used to adjust the system beauty settings based on preset beauty parameters to obtain initial beauty settings; An initial processing module, used for processing the live video stream based on the initial beauty setting to obtain an initial video stream; An initial push module, used to push the initial video stream to a third-party platform; Real-time monitoring module, used to monitor the changes of subsequent video streams in real time; A current beauty parameter acquisition module, used to acquire current beauty parameters when it is detected that the change of the subsequent video stream reaches a preset condition; A current beauty parameter setting module, used to adjust the system beauty settings based on the current beauty parameters to obtain the current beauty settings; A current processing module, used for processing the subsequent video stream based on the current beauty setting to obtain a current video stream; A current push module, used to push the current video stream to the third-party platform; The loop module is used to repeatedly execute the real-time monitoring module in sequence, to the current push module, until the live broadcast ends.

9. An electronic device, characterized in that: It includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement a method for automatically adjusting the beauty of a live broadcast by an anchor as described in any one of claims 1 to 7.

10. A computer-readable storage medium, wherein at least one instruction or at least one program is stored in the storage medium, and the at least one instruction or the at least one program is loaded and executed by a processor to implement a method for automatically adjusting the beauty of a live broadcast of an anchor as described in any one of claims 1 to 7.

Images