CN-114519889-B - Cover image detection method and device for live broadcasting room, computer equipment and medium

CN114519889BCN 114519889 BCN114519889 BCN 114519889BCN-114519889-B

Abstract

The application relates to the technical field of network live broadcasting and provides a cover image detection method, a device, computer equipment and a storage medium of a live broadcasting room, wherein the method comprises the steps of obtaining a cover image to be detected, carrying out face detection on the cover image to obtain a face detection result, obtaining the area size of each face area and the size of the cover image if one or more face areas exist in the face detection result, and carrying out face contour key point detection on each face area to obtain a face contour key point detection result if the ratio of each face area in the cover image is smaller than a preset ratio threshold value and the aspect ratio example is in a preset ratio threshold value range; if the detection results of the key points of the human face outline meet the preset conditions, the cover image is used as the target cover image, and the cover image detection efficiency is improved.

Inventors

WANG PU
CHEN ZENGHAI
ZHENG KANGYUAN

Assignees

广州方硅信息技术有限公司
广州方硅信息技术有限公司

Dates

Publication Date: 20260421
Application Date: 20220223
Priority Date: 20220223

Claims (9)

1. A cover image detection method for a live broadcast room, the method comprising the steps of: acquiring a cover image to be detected, wherein the cover image to be detected is a cover image obtained from live broadcast real-time screenshot or historical screenshot of live broadcast by a live broadcast platform; performing face detection on the cover image to obtain a face detection result; If one or more face areas exist in the face detection result, carrying out virtual human image detection on each face area to obtain a virtual human image detection result of each face area; if the virtual portrait detection result of each face area does not exist, acquiring the area size of each face area and the size of the cover image; If the ratio of the width of each face area to the width of the cover image, the ratio of the height of each face area to the height of the cover image and the ratio of the width to the height of each face area are calculated according to the area size of each face area and the size of the cover image, the ratio of each face area to the width of the cover image and the ratio of the width to the height of each face area are judged to be smaller than a preset ratio threshold value, the aspect ratio of each face area is within the preset ratio threshold value, and face contour key point detection is carried out on each face area to obtain a face contour key point detection result of each face area; And if the detection result of the key points of the face outline of each face area meets the preset condition, taking the cover image as a target cover image.
2. The cover image detection method of a living room according to claim 1, wherein: The step of performing face detection on the cover image to obtain a face detection result comprises the following steps: performing face detection on the cover image by using a first convolutional neural network to obtain a plurality of face detection frames including faces in the cover image and corresponding confidence scores; sequencing the face detection frames according to the sequence from high confidence score to low confidence score to obtain a face detection frame list; adding the face detection frame with the highest confidence score into an output list, and deleting the face detection frame with the highest confidence score from the face detection frame list; Calculating the overlapping degree of the face detection frame with the highest confidence degree score and other face detection frames, and deleting the face detection frames with the overlapping degree larger than the overlapping degree threshold value from the face detection frame list; And calculating the overlapping degree of the face detection frames with the highest confidence scores and other face detection frames remained in the face detection frame list, deleting the face detection frames with the overlapping degree larger than the overlapping degree threshold value from the face detection frame list until the face detection frame list is empty, and obtaining a face detection result according to the output list.
3. The cover image detection method of a living room according to claim 1, wherein: If one or more face areas exist in the face detection result, performing virtual image detection on each face area to obtain a virtual image detection result of each face area, including: If one or more face areas exist in the face detection result, performing virtual image detection on each face area by using a second convolutional neural network to obtain a virtual image confidence score of each face area; And if the confidence score of the virtual portrait is smaller than a preset score, determining that the corresponding face area does not have the virtual portrait.
4. The cover image detection method of a living room according to claim 1, wherein: if the ratio of each face region in the cover image is smaller than a preset ratio threshold and the aspect ratio of each face region is within a preset ratio threshold according to the region size of each face region and the size of the cover image, detecting face contour key points of each face region to obtain a face contour key point detection result of each face region, including: If the occupation ratio of each face region in the cover image is smaller than a preset occupation ratio threshold value and the aspect ratio of each face region is in a preset proportion threshold value range according to the region size of each face region and the size of the cover image, expanding the face regions outwards by a preset proportion along the diagonal direction of the face regions to obtain corresponding target face regions; Inputting each target face region into a face contour key point detection model to obtain a face contour key point detection result of each target face region, wherein the face contour key point detection result is used for indicating the position and confidence score of the face contour key point in each target face region.
5. The cover image detection method of a living room according to claim 4, wherein: and if the detection result of the key points of the face contour of each face area meets the preset condition, taking the cover image as a target cover image, wherein the step comprises the following steps: If the position of the face contour key point of each face area is in the cover image and the confidence score of the face contour key point of each face area is larger than a preset threshold, determining the face contour key point as a target contour key point; and if the number of the target contour key points is larger than a preset value, taking the cover image as a target cover image.
6.A cover image detection device for a live broadcast room, comprising: the cover image acquisition module is used for acquiring a cover image to be detected, wherein the cover image to be detected is a cover image obtained from live broadcast or history live broadcast of a live broadcast platform; The face detection module is used for carrying out face detection on the cover image to obtain a face detection result; The size acquisition module is used for carrying out virtual human image detection on each human face region if one or more human face regions exist in the human face detection result, so as to obtain a virtual human image detection result of each human face region; The key point detection module is used for calculating the ratio of the width of each face area to the width of the cover image, the ratio of the height of each face area to the height of the cover image and the ratio of the width of each face area to the height of each face area if the area size of each face area and the size of the cover image are calculated, judging that the ratio of each face area in the cover image is smaller than a preset ratio threshold value and the aspect ratio of each face area is in a preset ratio threshold value range, and detecting the key points of the face contour of each face area to obtain the key point detection result of the face contour of each face area; and the target cover image obtaining module is used for taking the cover image as a target cover image if the detection result of the key points of the human face contour of each human face area meets the preset condition.
7. The cover image detection apparatus of a living room as claimed in claim 6, wherein the face detection module includes: The face detection frame obtaining unit is used for carrying out face detection on the cover image by utilizing a first convolution neural network to obtain a plurality of face detection frames comprising faces and corresponding confidence scores in the cover image; a face detection frame list obtaining unit, configured to sort the face detection frames according to the order of the confidence scores from high to low, to obtain a face detection frame list; An output list adding unit, configured to add a face detection frame with the highest confidence score to an output list, and delete the face detection frame with the highest confidence score from the face detection frame list; The overlapping degree calculating unit is used for calculating the overlapping degree of the face detection frame with the highest confidence degree score and other face detection frames, and deleting the face detection frames with the overlapping degree larger than the overlapping degree threshold value from the face detection frame list; And the human face detection result obtaining unit is used for calculating the overlapping degree of the human face detection frame with the highest confidence score and other human face detection frames remained in the human face detection frame list, deleting the human face detection frame with the overlapping degree larger than the overlapping degree threshold value from the human face detection frame list until the human face detection frame list is empty, and obtaining the human face detection result according to the output list.
8. Computer device comprising a processor, a memory and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any one of claims 1 to 5 when the computer program is executed.
9. A computer readable storage medium storing a computer program, characterized in that the computer program when executed by a processor implements the steps of the method according to any one of claims 1 to 5.

Description

Cover image detection method and device for live broadcasting room, computer equipment and medium Technical Field The embodiment of the application relates to the technical field of network live broadcasting, in particular to a cover image detection method, a cover image detection device, computer equipment and a storage medium of a live broadcasting room. Background With the rapid development of internet technology, network live broadcast is becoming an entertainment means that is becoming popular. When a host is live, it is often necessary to use a live cover to present and recommend content of a different live to attract users' clicks and views. Specifically, after a user opens a live platform, such as a live application APP, a live list is presented on an interface of the live application APP, and each live content in the live list can be displayed through a live cover preset by a host, and the user clicks the corresponding live cover to enter a live room, so that the actual live content can be seen. At present, live covers mainly take live real-time screenshot or the most relevant picture of live history screenshot candidate set as a cover image, and the cover image obtained in the mode cannot guarantee the cover quality and needs to be checked one by one manually. However, the manual auditing cost is high and the efficiency is low facing the massive live covers. Disclosure of Invention The embodiment of the application provides a cover image detection method, a cover image detection device, computer equipment and a storage medium for a live broadcast room, which can solve the technical problems of low quality of the live broadcast cover, high cost and low efficiency of manual auditing of the live broadcast cover, and the technical scheme is as follows: In a first aspect, an embodiment of the present application provides a cover image detection method for a live broadcast room, including the steps of: Acquiring a cover image to be detected; performing face detection on the cover image to obtain a face detection result; If one or more face areas exist in the face detection result, acquiring the area size of each face area and the size of the cover image; If the occupation ratio of each face area in the cover image is smaller than a preset occupation ratio threshold value according to the area size of each face area and the size of the cover image, and the aspect ratio of each face area is in a preset proportion threshold value range, detecting face contour key points of each face area to obtain a face contour key point detection result of each face area; And if the detection result of the key points of the face outline of each face area meets the preset condition, taking the cover image as a target cover image. In a second aspect, an embodiment of the present application provides a cover image detection apparatus for a live broadcast room, including: The cover image acquisition module is used for acquiring a cover image to be detected; The face detection module is used for carrying out face detection on the cover image to obtain a face detection result; the size acquisition module is used for acquiring the area size of each face area and the size of the cover image if one or more face areas exist in the face detection result; The key point detection module is used for judging that the duty ratio of each face area in the cover image is smaller than a preset duty ratio threshold value according to the area size of each face area and the size of the cover image, and the aspect ratio of each face area is in a preset ratio threshold value range, carrying out face contour key point detection on each face area to obtain a face contour key point detection result of each face area; and the target cover image obtaining module is used for taking the cover image as a target cover image if the detection result of the key points of the human face contour of each human face area meets the preset condition. In a third aspect, embodiments of the present application provide a computer device, a processor, a memory and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the method as in the first aspect when the computer program is executed. In a fourth aspect, embodiments of the present application provide a computer readable storage medium storing a computer program which, when executed by a processor, performs steps as the method of the first aspect. The method comprises the steps of obtaining a cover image to be detected, carrying out face detection on the cover image to obtain a face detection result, obtaining the area size of each face area and the size of the cover image if one or more face areas exist in the face detection result, judging that the ratio of each face area in the cover image is smaller than a preset ratio threshold value according to the area size of each face area and the size of the cover image, carrying out face contour key point detec