System for identifying the amount of rouge used in the makeup of the live streamer's face

Abstract

The present invention relates to a system for identifying the usage amount of rouge on the face of an anchor. The system includes: a usage amount identification device, which is used to intelligently judge the predicted usage amount of the rouge to be applied for the anchor to apply makeup to the standard makeup before each live broadcast by using an intelligent usage amount analysis model based on the set red channel value, the area of rouge application in the standard makeup applied by the anchor before each live broadcast, and the curvature values corresponding to each point evenly spaced on the contour edge of the set reference facial contour pattern of the anchor; a wireless communication device, which is used to wirelessly transmit the predicted usage amount. Through this system, it is possible to intelligently predict the amount of rouge that needs to be consumed for the anchor to reach the standard makeup before each live broadcast by using an intelligent usage amount analysis model with a customized structural design based on a number of targeted screened basic information, so as to complete a reliable prediction of the amount of rouge that needs to be consumed for different anchors to apply makeup each time when using different standard makeup looks.

Description

Technical Field

[0001] The present invention relates to the field of live broadcast management, and more specifically, to a system for identifying the amount of rouge used in the makeup of a live streamer's face. Background Art

[0002] Live streamers usually wear makeup during live broadcasts. For the same live streamer, there are different standard makeup looks available before going live. For different standard makeup looks, there are slight differences in the amount of rouge used. If the live streamer is fixed, these slight differences can also be fixed. However, live streamers may change in the same live broadcast room or different live broadcast rooms. For different live streamers, when using different standard makeup looks, it is difficult to predict the amount of rouge needed for each makeup application, resulting in uncertainty about how much specific rouge should be prepared for each makeup application. Summary of the Invention

[0003] To solve the above problems, the present invention provides a system for identifying the amount of rouge used in the makeup of a live streamer's face, which can customize an intelligent usage analysis model with a structural design for the intelligent prediction of the amount of rouge needed for a set live streamer to achieve the standard makeup look before each live broadcast. The intelligent usage analysis model is a deep neural network that completes a set number of learning actions. The value of the set number is proportional to the rouge application area in the standard makeup look for a set live streamer before each live broadcast, thereby providing artificial intelligence models with different structures for predicting the amount of rouge for different standard makeup looks. At the same time, for the intelligent prediction of the amount of rouge needed for a set live streamer to achieve the standard makeup look before each live broadcast, a sufficient and comprehensive number of basic data items are specifically screened. The multiple basic data items include the set red channel value, the rouge application area in the standard makeup look for a set live streamer before each live broadcast, and the respective color channel values corresponding to the set rouge application color, the respective curvature values corresponding to evenly spaced points on the contour edge of the set live streamer's reference facial contour pattern, the area ratio of the skin in the reference facial contour pattern, and the respective depth of field values corresponding to each pixel point in the reference facial contour pattern. Among them, the respective color channel values corresponding to the set rouge application color in the standard makeup look for a set live streamer before each live broadcast are the red-green channel value, the black-white channel value, and the yellow-blue channel value of the set rouge application color in the LAB color space, and the rouge application area in the standard makeup look for a set live streamer before each live broadcast is the actual area of the live streamer's face where rouge is applied, thereby completing a reliable prediction of the amount of rouge needed for each makeup application when different live streamers use different standard makeup looks.

[0004] According to the present invention, there is provided a system for identifying the amount of rouge used in the makeup of a live streamer's face, the system comprising:

[0005] A first capture mechanism for obtaining the area of rouge application and the respective color channel values corresponding to the set rouge application color in the standard makeup for the set host before each live broadcast, where the rouge has a set red channel value;

[0006] A second capture mechanism for obtaining the respective curvature values corresponding to evenly spaced points on the contour edge of the set host's reference facial contour pattern, the area ratio of the skin in the reference facial contour pattern, and the respective depth of field values corresponding to each pixel point in the reference facial contour pattern;

[0007] A network construction mechanism for performing a set number of learning actions on a deep neural network to obtain a deep neural network that has completed the set number of learning actions and output it as an intelligent usage analysis model;

[0008] A usage identification device, connected to the first capture mechanism, the second capture mechanism, and the network construction mechanism respectively, for using the intelligent usage analysis model to intelligently judge the predicted usage value of the rouge to be applied for the set host to apply makeup to the standard makeup before each live broadcast based on the set red channel value, the area of rouge application in the standard makeup for the set host before each live broadcast, the respective color channel values corresponding to the set rouge application color, the respective curvature values corresponding to evenly spaced points on the contour edge of the set host's reference facial contour pattern, the area ratio of the skin in the reference facial contour pattern, and the respective depth of field values corresponding to each pixel point in the reference facial contour pattern;

[0009] A wireless communication device, connected to the usage identification device, for wirelessly transmitting the received predicted usage value to the portable terminal of the nearest live broadcast staff;

[0010] Among them, the network construction mechanism for performing a set number of learning actions on a deep neural network to obtain a deep neural network that has completed the set number of learning actions and output it as an intelligent usage analysis model includes: the value of the set number is proportional to the area of rouge application in the standard makeup for the set host before each live broadcast;

[0011] Among them, obtaining the area of rouge application and the respective color channel values corresponding to the set rouge application color in the standard makeup for the set host before each live broadcast includes: the respective color channel values corresponding to the set rouge application color in the standard makeup for the set host before each live broadcast are the red - green channel values, black - white channel values, and yellow - blue channel values of the set rouge application color in the LAB color space.

[0012] The logic of the system for identifying the rouge usage of the host's facial makeup in the present invention is reliable and the operation is intelligent. Since the intelligent usage analysis model with a customized structure design can be adopted to intelligently predict and set the rouge usage required for the host to achieve the standard makeup before each live broadcast based on a number of targeted basic information, the reliable prediction of the rouge usage required for each makeup application of different hosts when using different standard makeup can be completed. Brief Description of the Drawings

[0013] Those skilled in the art can better understand the numerous advantages of the present invention by referring to the accompanying drawings, where:

[0014] Figure 1 is a schematic structural diagram of the system for identifying the rouge usage of the host's facial makeup according to the primary embodiment of the present invention.

[0015] Figure 2 is a schematic structural diagram of the system for identifying the rouge usage of the host's facial makeup according to the secondary embodiment of the present invention.

[0016] Figure 3 is a schematic structural diagram of the system for identifying the rouge usage of the host's facial makeup according to the tertiary embodiment of the present invention. Detailed Description of the Invention

[0017] Figure 1 is a schematic structural diagram of the system for identifying the rouge usage of the host's facial makeup according to the primary embodiment of the present invention, and the system includes:

[0018] A first capture mechanism for obtaining the rouge application area in the standard makeup for the set host before each live broadcast and the respective color channel values corresponding to the set rouge application color, where the rouge has a set red channel value;

[0019] Specifically, the first capture mechanism for obtaining the rouge application area in the standard makeup for the set host before each live broadcast and the respective color channel values corresponding to the set rouge application color, where the rouge has a set red channel value includes: the first capture mechanism includes a color storage component for pre-storing the respective color channel values corresponding to the set rouge application color;

[0020] A second capture mechanism for obtaining the respective curvature values corresponding to evenly spaced points on the contour edge of the reference facial contour pattern of the set host, the area ratio of the skin occupied in the reference facial contour pattern, and the respective depth of field values corresponding to each pixel point in the reference facial contour pattern;

[0021] A network construction mechanism for performing a set number of learning actions on the deep neural network to obtain the deep neural network that has completed the set number of learning actions and output it as an intelligent usage analysis model;

[0022] A dosage identification device, which is respectively connected to the first capture mechanism, the second capture mechanism, and the network construction mechanism, and is used to use an intelligent dosage analysis model to intelligently judge the predicted dosage value of the rouge to be applied for the set host to apply makeup to the standard makeup before each live broadcast based on the set red channel value, the area of rouge application in the standard makeup before each live broadcast of the host, the respective color channel values corresponding to the set rouge application color, the respective curvature values corresponding to evenly spaced points on the contour edge of the set reference facial contour pattern of the host, the area ratio of the skin occupied in the reference facial contour pattern, and the respective depth of field values corresponding to each pixel point in the reference facial contour pattern;

[0023] A wireless communication device, which is connected to the dosage identification device and is used to wirelessly send the received dosage prediction value to the portable terminal of the nearest live broadcast staff;

[0024] Among them, the network construction mechanism is used to perform a set number of learning actions on the deep neural network to obtain a deep neural network that has completed a set number of learning actions, and outputs as an intelligent dosage analysis model including: the value of the set number is proportional to the area of rouge application in the standard makeup before each live broadcast of the host;

[0025] Among them, obtaining the area of rouge application in the standard makeup before each live broadcast of the host and the respective color channel values corresponding to the set rouge application color includes: the respective color channel values corresponding to the set rouge application color in the standard makeup before each live broadcast of the host are the red-green channel value, the black-white channel value, and the yellow-blue channel value of the set rouge application color in the LAB color space;

[0026] Among them, obtaining the area of rouge application in the standard makeup before each live broadcast of the host and the respective color channel values corresponding to the set rouge application color further includes: the area of rouge application in the standard makeup before each live broadcast of the host is the actual area of the host's face where rouge is applied;

[0027] Among them, the intelligent dosage analysis model is used to intelligently judge the predicted dosage value of the rouge to be applied when the host applies makeup to the standard makeup before each live broadcast, based on the set red channel value, the area of rouge application in the standard makeup before each live broadcast by the host, the respective color channel values corresponding to the set rouge application color, the respective curvature values corresponding to the evenly spaced points on the contour edge of the host's reference facial contour pattern, the area ratio of the skin occupied in the reference facial contour pattern, and the respective depth of field values corresponding to each pixel point in the reference facial contour pattern. The process includes: parallelly inputting the set red channel value, the area of rouge application in the standard makeup before each live broadcast by the host, the respective color channel values corresponding to the set rouge application color, the respective curvature values corresponding to the evenly spaced points on the contour edge of the host's reference facial contour pattern, the area ratio of the skin occupied in the reference facial contour pattern, and the respective depth of field values corresponding to each pixel point in the reference facial contour pattern into the intelligent dosage analysis model.

[0028] Figure 2 It is a schematic structural diagram of the rouge dosage identification system for the host's facial makeup according to the secondary embodiment of the present invention.

[0029] And Figure 1 different, Figure 2 the rouge dosage identification system for the host's facial makeup in

[0030] The current detection device is arranged near the first capture mechanism, the second capture mechanism, the network construction mechanism and the dosage identification device and is respectively connected to the first capture mechanism, the second capture mechanism, the network construction mechanism and the dosage identification device;

[0031] Among them, the current detection device is arranged near the first capture mechanism, the second capture mechanism, the network construction mechanism and the dosage identification device and is respectively connected to the first capture mechanism, the second capture mechanism, the network construction mechanism and the dosage identification device, including: the current detection device is used to respectively realize the on-site measurement of the current of the first capture mechanism, the second capture mechanism, the network construction mechanism and the dosage identification device.

[0032] Figure 3 It is a schematic structural diagram of the rouge dosage identification system for the host's facial makeup according to the secondary-secondary embodiment of the present invention.

[0033] And Figure 1 different, Figure 3 the rouge dosage identification system for the host's facial makeup in

[0034] A humidity detection device is arranged near the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device, and is respectively connected to the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device;

[0035] Among them, the humidity detection device arranged near the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device and respectively connected to the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device includes: the humidity detection device is used to respectively perform on-site measurement of the current humidity of the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device.

[0036] Next, the specific structure of the main anchor's facial makeup rouge dosage identification system of the present invention will be further described.

[0037] In the main anchor's facial makeup rouge dosage identification system according to various embodiments of the present invention:

[0038] An FPGA device is used to perform image data processing on the output data of the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device to obtain the output processed data corresponding to the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device respectively.

[0039] In the main anchor's facial makeup rouge dosage identification system according to various embodiments of the present invention:

[0040] The use of an FPGA device to perform image data processing on the output data of the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device to obtain the output processed data corresponding to the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device respectively includes: performing guided filtering processing on the output data of the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device to obtain the output processed data corresponding to the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device respectively.

[0041] In the main anchor's facial makeup rouge dosage identification system according to various embodiments of the present invention:

[0042] Using an FPGA device to perform image data processing on the output data of the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device to obtain the output processed data corresponding to the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device respectively includes: performing box filtering processing on the output data of the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device to obtain the output processed data corresponding to the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device respectively.

[0043] In the host face makeup rouge dosage identification system according to various embodiments of the present invention:

[0044] Using an FPGA device to perform image data processing on the output data of the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device to obtain the output processed data corresponding to the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device respectively includes: performing wavelet filtering processing on the output data of the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device to obtain the output processed data corresponding to the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device respectively.

[0045] And in the host face makeup rouge dosage identification system according to various embodiments of the present invention:

[0046] Using an FPGA device to perform image data processing on the output data of the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device to obtain the output processed data corresponding to the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device respectively includes: performing bilinear interpolation processing on the output data of the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device to obtain the output processed data corresponding to the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device respectively.

[0047] In addition, in the system for identifying the usage amount of rouge in the face makeup of the live streamer, the intelligent usage amount analysis model is used to intelligently judge the predicted usage amount of the rouge to be applied for the live streamer to apply makeup to the standard makeup before each live broadcast based on the set red channel value, the set rouge application area in the standard makeup applied by the live streamer before each live broadcast, the respective color channel values corresponding to the set rouge application color, the respective curvature values corresponding to the evenly spaced points on the contour edge of the set reference facial contour pattern of the live streamer, the area ratio of the skin in the reference facial contour pattern, and the respective depth of field values corresponding to each pixel point in the reference facial contour pattern. It further includes: executing the intelligent usage amount analysis model to obtain the predicted usage amount of the rouge to be applied for the live streamer to apply makeup to the standard makeup before each live broadcast output by the intelligent usage amount analysis model.

[0048] The preferred embodiments of the present invention have been discussed in detail so far, but these embodiments are only specific examples for clarifying the technical content of the present invention. Therefore, the present invention should not be considered limited to these specific examples. The spirit and scope of the present invention are only defined by the appended claims.

Claims

1. A system for identifying the amount of rouge used in the makeup of an anchor's face, characterized in that, The system includes: A first capture mechanism for obtaining the rouge application area in the standard makeup before each live broadcast of a set host and the respective color channel values corresponding to the set rouge application color, where the rouge has a set red channel value; A second capture mechanism for obtaining the respective curvature values corresponding to evenly spaced points on the contour edge of the reference facial contour pattern of the set host, the area ratio of the skin in the reference facial contour pattern, and the respective depth of field values corresponding to each pixel point in the reference facial contour pattern; A network construction mechanism for performing a set number of learning actions on a deep neural network to obtain a deep neural network that has completed the set number of learning actions and output it as an intelligent usage analysis model; A usage identification device, connected to the first capture mechanism, the second capture mechanism, and the network construction mechanism respectively, for using the intelligent usage analysis model to intelligently judge the predicted usage value of the rouge to be applied to the standard makeup before each live broadcast of the set host based on the set red channel value, the rouge application area in the standard makeup before each live broadcast of the set host, the respective color channel values corresponding to the set rouge application color, the respective curvature values corresponding to evenly spaced points on the contour edge of the reference facial contour pattern of the set host, the area ratio of the skin in the reference facial contour pattern, and the respective depth of field values corresponding to each pixel point in the reference facial contour pattern; A wireless communication device, connected to the usage identification device, for wirelessly sending the received predicted usage value to the portable terminal of the nearest live broadcast staff; Among them, the network construction mechanism for performing a set number of learning actions on a deep neural network to obtain a deep neural network that has completed the set number of learning actions and output it as an intelligent usage analysis model includes: the value of the set number is proportional to the rouge application area in the standard makeup before each live broadcast of the set host; Among them, obtaining the rouge application area in the standard makeup before each live broadcast of the set host and the respective color channel values corresponding to the set rouge application color includes: the respective color channel values corresponding to the set rouge application color in the standard makeup before each live broadcast of the set host are the red - green channel values, black - white channel values, and yellow - blue channel values of the set rouge application color in the LAB color space.

2. The system for identifying the rouge usage of the host's face makeup according to claim 1, wherein: Obtaining the rouge application area in the standard makeup before each live broadcast of the set host and the respective color channel values corresponding to the set rouge application color further includes: the rouge application area in the standard makeup before each live broadcast of the set host is the actual area of the host's face where the rouge is applied. Among them, the intelligent dosage analysis model is used to intelligently judge the predicted dosage value of the rouge to be applied when the host applies makeup to the standard makeup before each live broadcast based on the set red channel value, the area of rouge application in the standard makeup before each live broadcast of the host, the respective color channel values corresponding to the set rouge application color, the respective curvature values corresponding to evenly spaced points on the contour edge of the host's reference facial contour pattern, the area ratio of the skin occupied in the reference facial contour pattern, and the respective depth-of-field values corresponding to each pixel point in the reference facial contour pattern, including: inputting the set red channel value, the area of rouge application in the standard makeup before each live broadcast of the host, the respective color channel values corresponding to the set rouge application color, the respective curvature values corresponding to evenly spaced points on the contour edge of the host's reference facial contour pattern, the area ratio of the skin occupied in the reference facial contour pattern, and the respective depth-of-field values corresponding to each pixel point in the reference facial contour pattern into the intelligent dosage analysis model in parallel.

3. The host face makeup rouge usage identification system according to claim 2, characterized in that, The system further includes: Current detection devices, which are arranged near the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device and are respectively connected to the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device; Among them, the current detection devices, which are arranged near the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device and are respectively connected to the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device, include: the current detection devices are used to respectively perform on-site measurement of the current of the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device.

4. The system for identifying the rouge usage amount of the host's facial makeup according to claim 2, wherein, The system further includes: Humidity detection devices, which are arranged near the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device and are respectively connected to the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device; Among them, the humidity detection devices, which are arranged near the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device and are respectively connected to the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device, include: the humidity detection devices are used to respectively perform on-site measurement of the current humidity of the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device.

5. The host face makeup rouge dosage identification system according to any one of claims 2-4, wherein: An FPGA device is used to perform image data processing on the output data of the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device to obtain the output processed data corresponding to the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device respectively.

6. The system for identifying the rouge dosage of the host's facial makeup according to claim 5, wherein: Using an FPGA device to perform image data processing on the output data of the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device to obtain the output processed data corresponding to the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device respectively includes: performing guided filtering processing on the output data of the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device to obtain the output processed data corresponding to the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device respectively.

7. The system for identifying the rouge dosage of the host's facial makeup according to claim 5, wherein: Using an FPGA device to perform image data processing on the output data of the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device to obtain the output processed data corresponding to the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device respectively includes: performing box filtering processing on the output data of the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device to obtain the output processed data corresponding to the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device respectively.

8. The system for identifying the rouge dosage of the host's facial makeup according to claim 5, wherein: Using an FPGA device to perform image data processing on the output data of the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device to obtain the output processed data corresponding to the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device respectively includes: performing wavelet filtering processing on the output data of the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device to obtain the output processed data corresponding to the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device respectively.

9. The system for identifying the rouge dosage of the host's facial makeup according to claim 5, wherein: Using an FPGA device to perform image data processing on the output data of the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device to obtain the output processing data corresponding to the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device respectively includes: performing bilinear interpolation processing on the output data of the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device to obtain the output processing data corresponding to the first capture mechanism, the second capture mechanism, the network construction mechanism, and the dosage identification device respectively.

Images