A contact lens virtual try-on and intelligent makeup coordination recommendation system and method
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-28
- Publication Date
- 2026-08-11
AI Technical Summary
然而,这类系统大多只展示单一产品的效果,缺乏对整体妆容的协调性考虑,导致用户体验较为局限,尤其在妆容搭配的场景中,无法提供智能化的推荐和优化
Smart Images

Figure CN120447738B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of augmented reality technology, and in particular to a virtual try-on system and method for contact lenses and intelligent makeup recommendation. Background Technology
[0002] Currently, with the development of digital and intelligent technologies, virtual try-on systems have gradually entered daily life, especially in the beauty industry. Traditional virtual try-on systems typically focus on showcasing single products, such as contact lenses, lipsticks, and eyeshadows. Users can upload photos or use a live camera to see how these products look on their faces. However, most of these systems only display the effect of a single product and lack consideration for the overall coordination of the makeup, resulting in a limited user experience, especially in makeup matching scenarios, where they cannot provide intelligent recommendations and optimizations.
[0003] Therefore, a solution is urgently needed. Summary of the Invention
[0004] One objective of this invention is to provide a virtual contact lens try-on and intelligent makeup recommendation system. This system acquires the user's facial features and matching needs, automatically recommending suitable contact lenses and facial makeup based on these factors. Finally, it assists the user in virtually trying on contact lenses and virtually applying facial makeup, providing intelligent recommendations for contact lens wearing and makeup coordination. This considers the overall harmony of the user's makeup, thus improving the user experience.
[0005] This invention provides a virtual try-on system for contact lenses and a smart makeup recommendation system, comprising:
[0006] The acquisition module is used to acquire the user's facial features and matching requirements;
[0007] The recommendation module is used to recommend matching contact lenses and facial makeup to users based on facial features and matching needs;
[0008] The auxiliary module is used to help users virtually try on contact lenses and virtually apply facial makeup.
[0009] Optionally, the facial features include at least: face shape, eyebrow shape, eye size and shape, eye spacing, nose shape, lip shape, skin color, and facial proportions;
[0010] The matching requirements include at least: travel scenarios, preferred makeup styles, types of activities participated in, and functional requirements for contact lenses.
[0011] Optionally, the recommendation module recommends matching contact lenses and facial makeup to the user based on facial features and matching needs, including:
[0012] Based on a pre-trained collaborative matching model, the system recommends matching contact lenses and facial makeup to users according to their facial features and matching needs.
[0013] The pre-training steps for the collaborative matching model are as follows:
[0014] A large amount of matching information labeled with standard matching schemes is used as training samples for machine learning training to obtain a synergistic matching model.
[0015] Optionally, the auxiliary module assists the user in virtually trying on contact lenses and virtually applying facial makeup, including:
[0016] Separately plan the virtual application rhythm of facial makeup and the virtual timing of wearing contact lenses;
[0017] Based on the virtual face-applying rhythm, the interactive system assists users in applying facial makeup.
[0018] During the interactive process of assisting the user in applying facial makeup, when the virtual wearing opportunity arises, the interactive assisting user virtually tries on contact lenses.
[0019] Optionally, the separate planning of the virtual application rhythm of facial makeup and the virtual wearing timing of contact lenses includes:
[0020] Analyzing multiple makeup elements of facial makeup;
[0021] Based on the constraints of makeup element matching, various makeup elements are matched to obtain multiple makeup slices;
[0022] Based on the serialization processing constraints, each makeup slice is serialized to obtain a makeup slice sequence;
[0023] By imbuing the makeup slice sequence with a rhythm control mechanism, a virtual face-applying rhythm can be obtained;
[0024] The virtual wearing time for contact lenses is generated when the user's average actual acceptance of the virtual face-wearing of makeup slices for a consecutive threshold number of times exceeds the actual acceptance threshold.
[0025] The constraints on the combination of makeup elements include:
[0026] The same makeup element is contained in at least two makeup slices;
[0027] In addition, each makeup slice must contain at least one makeup element related to the eyes;
[0028] In addition, the similarity between any two makeup slices does not exceed the similarity threshold;
[0029] The serialization processing constraints include:
[0030] The user prediction preference and sequence order of each makeup slice in the makeup slice sequence are mapped into the curve coordinate system to obtain a complete peak;
[0031] The rhythm control mechanism includes:
[0032] Step 1: Apply each makeup slice in the makeup slice sequence to the face in the order they appear in the sequence;
[0033] Step 2: When the i-th makeup slice in the makeup slice sequence is virtually applied to the face, the value of j is determined based on the user's actual acceptance level at that time;
[0034] Step 3: Virtually apply the (i+j)th makeup slice in the makeup slice sequence to the face, and then use the (i+j)th makeup slice as the new ith makeup slice to return and repeat Step 2. This process is repeated.
[0035] Where 1≤i+j≤N, and N is the total number of makeup slices in the makeup slice sequence.
[0036] Optionally, the auxiliary module assists users in virtually trying on contact lenses and virtually applying facial makeup, and further includes:
[0037] When a user doesn't like the virtual contact lenses / virtual face makeup they're trying on, the system helps them quickly select a replacement contact lenses / makeup to try on / replace the virtual face.
[0038] The quick selection steps for replacing contact lenses / changing facial makeup are as follows:
[0039] Get the sequence of user operations within the most recent preset time period, such as zooming in on the target interface to view virtual contact lenses and virtual facial makeup.
[0040] The operation frequency of the parsing operation record sequence;
[0041] When the operation frequency does not exceed the frequency threshold, a first quick selection list is generated based on the first replacement material library of virtual contact lenses / virtual face makeup that the user does not like to try on, and the user can quickly select replacement contact lenses / replace face makeup from the first quick selection list;
[0042] Otherwise, the filter time period of the parsing operation record sequence is used; where the first and last times of the filter time period are the two times with the greatest distance between them among multiple times when the same area is viewed in local magnification for more than the dwell time threshold.
[0043] Parse the set of operation elements in the operation region of the operation record sequence within the filtered time period;
[0044] Based on the set of operation elements, match the second replacement material library;
[0045] Based on the second replacement material library, a second quick selection list is generated, allowing users to quickly select from the second quick selection list to replace contact lenses or facial makeup.
[0046] This invention provides a method for virtual try-on of contact lenses and intelligent makeup collaborative recommendation, comprising:
[0047] Obtain the user's facial features and matching preferences;
[0048] Based on facial features and matching needs, we recommend matching contact lenses and facial makeup for users;
[0049] It helps users virtually try on contact lenses and virtually apply facial makeup.
[0050] Optionally, the facial features include at least: face shape, eyebrow shape, eye size and shape, eye spacing, nose shape, lip shape, skin color, and facial proportions;
[0051] The matching requirements include at least: travel scenarios, preferred makeup styles, types of activities participated in, and functional requirements for contact lenses.
[0052] Optionally, the recommendation module recommends matching contact lenses and facial makeup to the user based on facial features and matching needs, including:
[0053] Based on a pre-trained collaborative matching model, the system recommends matching contact lenses and facial makeup to users according to their facial features and matching needs.
[0054] The pre-training steps for the collaborative matching model are as follows:
[0055] A large amount of matching information labeled with standard matching schemes is used as training samples for machine learning training to obtain a synergistic matching model.
[0056] Optionally, the auxiliary module assists the user in virtually trying on contact lenses and virtually applying facial makeup, including:
[0057] Separately plan the virtual application rhythm of facial makeup and the virtual timing of wearing contact lenses;
[0058] Based on the virtual face-applying rhythm, the interactive system assists users in applying facial makeup.
[0059] During the interactive process of assisting the user in applying facial makeup, when the virtual wearing opportunity arises, the interactive assisting user virtually tries on contact lenses.
[0060] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.
[0061] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description
[0062] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:
[0063] Figure 1 This is a schematic diagram of a virtual try-on system for contact lenses and a smart makeup recommendation system according to an embodiment of the present invention;
[0064] Figure 2 This is a schematic diagram of a method for virtual try-on of contact lenses and intelligent makeup recommendation in an embodiment of the present invention. Detailed Implementation
[0065] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.
[0066] This invention provides a virtual try-on system for contact lenses and a smart makeup recommendation system, such as... Figure 1 As shown, it includes:
[0067] Module 1 is used to acquire the user's facial features and matching requirements;
[0068] Recommendation module 2 is used to recommend matching contact lenses and facial makeup to users based on facial features and matching needs;
[0069] Auxiliary module 3 is used to assist users in virtually trying on contact lenses and virtually applying facial makeup.
[0070] The facial features include at least: face shape, eyebrow shape, eye size and shape, eye spacing, nose shape, lip shape, skin color, and facial proportions;
[0071] The matching requirements include at least: travel scenarios, preferred makeup styles, types of activities participated in, and functional requirements for contact lenses.
[0072] Facial features can be captured by taking photos using the camera on the user's smart device. Matching preferences can be input by the user through their smart device, which can be a mobile phone or a service device placed in a public place. Based on facial features and matching preferences, the system automatically recommends suitable contact lenses and makeup looks for the user, then assists the user in virtually trying on the contact lenses and virtually applying the makeup.
[0073] For example, a female user is about to attend a formal ball and has the following characteristics and needs:
[0074] Face shape: oval
[0075] Skin tone: Warm (with a slight golden tint)
[0076] Eyes: Large, almond-shaped
[0077] Matching requirements: Prefers an elegant and romantic makeup style, and wants to enhance the brightness and depth of the eyes when wearing contact lenses.
[0078] The ball is a formal occasion, so she needs a sophisticated and glamorous makeup look and contact lenses. Based on her face shape, the system suggests makeup that accentuates her facial contours, such as soft blush and appropriate highlighting. Her warm skin tone suits gold or copper-toned eyeshadow and a warm lip color; the system recommends this makeup color scheme to enhance her complexion. Because her large, almond-shaped eyes are suitable for some deep eye makeup, the system recommends smoky eyes or metallic eyeshadow, along with mascara to enlarge the eyes and make them look brighter and more expressive. Since the user wants to increase the brightness and depth of her eyes, the system recommends a pair of dark brown or purple contact lenses. After the user selects the recommended contact lenses and makeup style, the system uses augmented reality technology to let the user see the effect of wearing the contact lenses and makeup. She can observe in real time how the contact lenses make her eyes look deeper and how the makeup complements her face shape and skin tone.
[0079] This application obtains the user's facial features and matching needs, and automatically recommends matching contact lenses and facial makeup based on these two factors. Finally, it assists the user in virtually trying on contact lenses and virtually applying facial makeup, providing the user with intelligent contact lens wearing and makeup matching recommendations, taking into account the overall coordination of the user's makeup, and improving the user experience.
[0080] In one embodiment, the recommendation module recommends matching contact lenses and facial makeup to the user based on facial features and matching needs, including:
[0081] Based on a pre-trained collaborative matching model, the system recommends matching contact lenses and facial makeup to users according to their facial features and matching needs.
[0082] The pre-training steps for the collaborative matching model are as follows:
[0083] A large amount of matching information labeled with standard matching schemes is used as training samples for machine learning training to obtain a synergistic matching model.
[0084] The standard matching scheme is a combination of contact lenses and facial makeup pre-designed by stylists based on matching criteria. These criteria include individual facial features and makeup preferences. By using a large amount of matching criteria labeled with the standard scheme as training samples for machine learning training, the resulting collaborative matching model can adaptively recommend contact lenses and facial makeup to users based on facial features and makeup preferences, thus improving the efficiency of contact lens and facial makeup recommendations.
[0085] Specifically, most current virtual makeup try-on technologies project pre-set makeup effects directly onto the user's face. These makeup looks are usually standard effects set by designers or programs, covering different makeup styles, tones, and combinations. However, because these makeup looks are often quite complex or overly intense, many users may feel uncomfortable upon first seeing them, or find the effects too exaggerated, leading to a rejection. For example, users who prefer natural makeup might immediately abandon the try-on experience or even close the virtual try-on interface if they see a heavy smoky eye or bright red lipstick, thus wasting the opportunity and negatively impacting the user experience.
[0086] Furthermore, the timing of contact lens fitting is indeed crucial in virtual try-on technology, as it directly impacts the user experience, especially the way it complements the makeup look. If the contact lenses are put on immediately at the beginning, it might interfere with the user's overall perception of the makeup.
[0087] To address the aforementioned technical problems, in one embodiment, the auxiliary module assists the user in virtually trying on contact lenses and virtually applying facial makeup, including:
[0088] Separately plan the virtual application rhythm of facial makeup and the virtual timing of wearing contact lenses;
[0089] Based on the virtual face-applying rhythm, the interactive system assists users in applying facial makeup.
[0090] During the interactive process of assisting the user in applying facial makeup, when the virtual wearing opportunity arises, the interactive assisting user virtually tries on contact lenses.
[0091] The virtual face-applying rhythm refers to the rhythm of applying facial makeup in layers; the virtual wearing timing refers to the appropriate time for the user to virtually wear contact lenses while applying facial makeup; after these are planned, based on the virtual face-applying rhythm, the interactive system assists the user in applying facial makeup, and when the virtual wearing timing is reached, the interactive system assists the user in virtually trying on contact lenses.
[0092] This invention provides an interactive, face-applying assistance system that assists users in applying facial makeup according to a virtual face-applying rhythm. This avoids the discomfort or other adverse feelings that might result from directly applying makeup to the user's face virtually, and also avoids wasting the opportunity for virtual try-on, thus improving the user experience. Secondly, during the interactive application of facial makeup, when the virtual wearing opportunity arises, the system assists the user in virtually trying on contact lenses. This avoids directly trying on contact lenses, which could interfere with the user's overall perception of the makeup, further enhancing the user experience.
[0093] In one embodiment, the separate planning of the virtual application rhythm of facial makeup and the virtual wearing timing of contact lenses includes:
[0094] This section analyzes various elements of facial makeup; makeup elements include at least: base makeup, eye makeup, lip makeup, contouring and highlighting, and decorative items.
[0095] Based on the constraints of makeup element matching, various makeup elements are matched to obtain multiple makeup slices; under the constraints of serialization processing, the multiple sets of matched makeup elements each form different makeup slices.
[0096] Based on the serialization processing constraints, each makeup slice is serialized to obtain a makeup slice sequence; under the constraints of the serialization processing, each makeup slice is sorted to form a makeup slice sequence.
[0097] By imbuing the makeup slice sequence with a rhythm control mechanism, a virtual face-applying rhythm is obtained; the makeup slice sequence indicates the makeup slices that need to be virtually applied to the user's face in sequence, and by imbuing them with a rhythm control mechanism, a virtual face-applying rhythm for facial makeup is formed.
[0098] The virtual wearing opportunity for contact lenses is generated based on the following criteria: the user's average actual acceptance of virtual face-wearing of makeup slices for a consecutive threshold number of times exceeds the actual acceptance threshold; after each virtual face-wearing of a makeup slice, an actual acceptance feedback form is pushed to the user for them to fill in the actual acceptance, which represents the user's degree of acceptance of the makeup slice used for virtual face-wearing; the threshold can be 4; the actual acceptance threshold represents the threshold that represents the user's relatively high degree of acceptance of the makeup slice used for virtual face-wearing; when the user's average actual acceptance of virtual face-wearing of makeup slices for a consecutive threshold number of times exceeds the actual acceptance threshold, it indicates that the user has a relatively good understanding of the overall makeup and can perform virtual face-wearing of contact lenses, thus entering the virtual wearing opportunity;
[0099] The constraints on the combination of makeup elements include:
[0100] The same makeup element is contained in at least two makeup slices;
[0101] In addition, each makeup slice must contain at least one eye-related makeup element; eye-related makeup elements must include at least: eyeshadow, eyeliner, eyebrows, eye primer, etc.
[0102] Furthermore, the similarity between any two makeup slices must not exceed a similarity threshold; the similarity threshold represents a relatively high degree of similarity between two makeup slices. First, ensure that each makeup element appears in at least two makeup slices. This ensures a more balanced and harmonious application of makeup elements, preventing any elements from appearing too abrupt or unbalanced. Second, each makeup slice must contain at least one eye-related makeup element, ensuring that the eyes always occupy a prominent position in the makeup slice, thus making the overall design visually more focused on the eyes and matching the virtual wearing effect of contact lenses. Finally, by limiting the similarity between makeup slices, the uniqueness and creativity of each slice are enhanced, ensuring the diversity and personalization of makeup styles. Overall, these constraints help to create a makeup slice that is both balanced and aesthetically pleasing, as well as full of variation and creativity.
[0103] The serialization processing constraints include:
[0104] In a makeup slice sequence, the user's predicted preference for each makeup slice and its sequence order are mapped onto a curve coordinate system, forming a complete peak. The horizontal axis of the curve coordinate system represents the sequence order, and the vertical axis represents the predicted preference. Mapping each makeup slice's predicted preference and sequence order onto the curve coordinate system yields a coordinate point. Connecting these points sequentially creates the curve. If the curve forms a peak, it indicates that when each makeup slice in the sequence is virtually applied to the face in sequence, the user gradually adapts and accepts it (the rising phase of the complete peak), reaching the highest level of acceptance (the peak of the complete peak). Then, they gradually try makeup styles that might be less acceptable (the falling phase of the complete peak), greatly improving the user experience. The user's predicted preference can be calculated as the average selection frequency of different makeup element pairs from the makeup slices in the user's history.
[0105] The rhythm control mechanism includes:
[0106] Step 1: Apply each makeup slice in the makeup slice sequence to the face in the order they appear in the sequence;
[0107] Step 2: When the i-th makeup slice in the makeup slice sequence is virtually applied to the face, the value of j is determined based on the user's actual acceptance level at that time;
[0108] Step 3: Virtually apply the (i+j)th makeup slice in the makeup slice sequence to the face, and then use the (i+j)th makeup slice as the new ith makeup slice to return and repeat Step 2. This process is repeated.
[0109] Where 1≤i+j≤N, and N is the total number of makeup slices in the makeup slice sequence.
[0110] The j-value is positively correlated with the user's actual acceptance level in a given instance. A higher actual acceptance level allows for a greater leap in the next virtual makeup look segment, resulting in a higher j-value. Specifically, technical staff can pre-set a lookup table containing j-values corresponding to different users' actual acceptance levels. This table is consulted each time the j-value is determined. By evaluating the user's actual acceptance level for each virtual makeup look segment in real time, the system dynamically adjusts the order of subsequent makeup displays. Specifically, based on user feedback, the system calculates the j-value to determine how many makeup segments to skip before displaying the next most suitable one for virtual makeup. When actual acceptance is high, the j-value increases, allowing the system to skip more makeup segments and quickly display other styles. When acceptance is low, the j-value is smaller, maintaining the virtual makeup look order as much as possible and avoiding rapid switching. This makes the virtual makeup try-on process more personalized and interactive, improving the user experience, avoiding redundant displays, and greatly optimizing the pace and effect of virtual makeup recommendations.
[0111] In one embodiment, the auxiliary module assists the user in virtually trying on contact lenses and virtually applying facial makeup, and further includes:
[0112] When a user doesn't like the virtual contact lenses / virtual face makeup they're trying on, the system helps them quickly select a replacement contact lenses / makeup to try on / replace the virtual face.
[0113] The quick selection steps for replacing contact lenses / changing facial makeup are as follows:
[0114] Get the sequence of operation records of users zooming in on the target interface that displays virtual try-on contact lenses and virtual facial makeup within the most recent preset time; the most recent preset time can be the most recent 100 seconds; the operation record sequence is the operation record of zooming in on the target interface in chronological order of the time of generation, and the operation record includes the operation time, zoomed-in area, etc.
[0115] The operation frequency is analyzed from the operation record sequence; based on the operation record sequence, the frequency with which the user zooms in on a specific part of the target interface within the most recent preset time period can be determined.
[0116] When the operation frequency does not exceed the frequency threshold, a first quick selection list is generated based on the first replacement material library of virtual try-on contact lenses / virtual face makeup that the user dislikes. The user can then quickly select from the first quick selection list to replace the contact lenses / face makeup. The frequency threshold represents a threshold indicating a high viewing frequency. Generally, if a user dislikes the virtual try-on contact lenses / virtual face makeup completely, they will not zoom in to view it on the target interface. Therefore, when the operation frequency does not exceed the frequency threshold, it means that the user basically dislikes the virtual try-on contact lenses / virtual face makeup completely. The first quick selection list is then generated directly based on the first replacement material library of virtual try-on contact lenses / virtual face makeup that the user dislikes. The user can then quickly select from the first quick selection list to replace the contact lenses / face makeup. The first replacement material library contains a large number of materials that can replace the virtual try-on contact lenses / virtual face makeup that the user dislikes.
[0117] Otherwise, analyze the filtering time period of the operation record sequence; where the first and last moments of the filtering time period are the two farthest moments among multiple moments when the user stays to zoom in on the same area for more than the dwell time threshold; otherwise, it means that the user is still interested in a certain part of the virtual try-on contact lenses / virtual face makeup; the dwell time threshold is the threshold representing the longer dwell time for zooming in on the area; by taking the two farthest moments among multiple moments when the user stays to zoom in on the same area for more than the dwell time threshold as the first and last moments of the filtering time period, the more likely it is to find the part that the user is interested in within the filtering time period;
[0118] The operation element set of the operation record sequence within the selected time period is parsed; the operation element set includes makeup elements, contact lens elements, etc. in the operation area;
[0119] Based on the set of operation elements, a second replacement material library is matched; the second replacement material library contains materials that are related to the set of operation elements and are different from the virtual contact lenses / virtual face makeup that the user does not like to try on.
[0120] Based on the second replacement material library, a second quick selection list is generated, allowing users to quickly select from the second quick selection list to replace contact lenses or facial makeup.
[0121] This invention provides rapid and accurate personalized recommendations to help users find contact lenses or makeup looks that better suit their needs, reducing unnecessary operations and waiting time, and enhancing the accuracy and satisfaction of virtual try-on. Based on the operation frequency of the operation record sequence, it quickly determines whether the user fundamentally dislikes the virtual contact lenses / makeup looks and provides corresponding quick selection assistance to replace them, greatly improving the accuracy of assisting users in quickly selecting replacement contact lenses / makeup looks and enhancing the system's applicability.
[0122] This invention provides a method for virtual try-on of contact lenses and intelligent makeup collaborative recommendation, such as... Figure 2 As shown, it includes:
[0123] S1. Obtain the user's facial features and matching requirements;
[0124] S2. Based on facial features and matching needs, recommend matching contact lenses and facial makeup for users;
[0125] S3 assists users in virtually trying on contact lenses and virtually applying facial makeup.
[0126] The facial features include at least: face shape, eyebrow shape, eye size and shape, distance between eyes, nose shape, lip shape, skin color, and facial proportions;
[0127] The matching requirements include at least: travel scenarios, preferred makeup styles, types of activities participated in, and functional requirements for contact lenses.
[0128] The recommendation module recommends matching contact lenses and facial makeup based on facial features and matching needs, including:
[0129] Based on a pre-trained collaborative matching model, the system recommends matching contact lenses and facial makeup to users according to their facial features and matching needs.
[0130] The pre-training steps for the collaborative matching model are as follows:
[0131] A large amount of matching information labeled with standard matching schemes is used as training samples for machine learning training to obtain a synergistic matching model.
[0132] The auxiliary module assists users in virtually trying on contact lenses and virtually applying facial makeup, including:
[0133] Separately plan the virtual application rhythm of facial makeup and the virtual timing of wearing contact lenses;
[0134] Based on the virtual face-applying rhythm, the interactive system assists users in applying facial makeup.
[0135] During the interactive process of assisting the user in applying facial makeup, when the virtual wearing opportunity arises, the interactive assisting user virtually tries on contact lenses.
[0136] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.
Claims
1. A virtual try-on system for contact lenses and a smart makeup recommendation system, characterized in that, include: The acquisition module is used to acquire the user's facial features and matching requirements; The recommendation module is used to recommend matching contact lenses and facial makeup to users based on facial features and matching needs; The auxiliary module is used to help users virtually try on contact lenses and virtually apply facial makeup. The auxiliary module assists users in virtually trying on contact lenses and virtually applying facial makeup, including: Separately plan the virtual application rhythm of facial makeup and the virtual timing of wearing contact lenses; Based on the virtual face-applying rhythm, the interactive system assists users in applying facial makeup. During the interactive process of assisting the user in applying facial makeup, when the virtual wearing opportunity arrives, the interactive user virtually tries on contact lenses. The separately planned virtual application rhythm of facial makeup and virtual wearing timing of contact lenses include: Analyzing multiple makeup elements of facial makeup; Based on the constraints of makeup element matching, various makeup elements are matched to obtain multiple makeup slices; Based on the serialization processing constraints, each makeup slice is serialized to obtain a makeup slice sequence; By imbuing the makeup slice sequence with a rhythm control mechanism, a virtual face-applying rhythm can be obtained; The virtual wearing time for contact lenses is generated when the user's average actual acceptance of the virtual face-wearing of makeup slices for a consecutive threshold number of times exceeds the actual acceptance threshold. The constraints on the combination of makeup elements include: The same makeup element is contained in at least two makeup slices; In addition, each makeup slice must contain at least one makeup element related to the eyes; In addition, the similarity between any two makeup slices does not exceed the similarity threshold; The serialization processing constraints include: The user prediction preference and sequence order of each makeup slice in the makeup slice sequence are mapped into the curve coordinate system to obtain a complete peak; The rhythm control mechanism includes: Step 1: Apply each makeup slice in the makeup slice sequence to the face in the order they appear in the sequence; Step 2: When the i-th makeup slice in the makeup slice sequence is virtually applied to the face, the value of j is determined based on the user's actual acceptance level at that time; Step 3: Virtually apply the (i+j)th makeup slice in the makeup slice sequence to the face, and then use the (i+j)th makeup slice as the new ith makeup slice to return and repeat Step 2. This process is repeated. Where 1≤i+j≤N, and N is the total number of makeup slices in the makeup slice sequence.
2. The contact lens virtual try-on and intelligent makeup collaborative recommendation system as described in claim 1, characterized in that, The facial features include at least: face shape, eyebrow shape, eye size and shape, distance between eyes, nose shape, lip shape, skin color, and facial proportions; The matching requirements include at least: travel scenarios, preferred makeup styles, types of activities participated in, and functional requirements for contact lenses.
3. The contact lens virtual try-on and intelligent makeup collaborative recommendation system as described in claim 1, characterized in that, The recommendation module recommends matching contact lenses and facial makeup based on facial features and matching needs, including: Based on a pre-trained collaborative matching model, the system recommends matching contact lenses and facial makeup to users according to their facial features and matching needs. The pre-training steps for the collaborative matching model are as follows: A large amount of matching information labeled with standard matching schemes is used as training samples for machine learning training to obtain a synergistic matching model.
4. The contact lens virtual try-on and intelligent makeup collaborative recommendation system as described in claim 1, characterized in that, The auxiliary module assists users in virtually trying on contact lenses and virtually applying facial makeup, and also includes: When a user doesn't like the virtual contact lenses / virtual face makeup they're trying on, the system helps them quickly select a replacement contact lenses / makeup to try on / replace the virtual face. The quick selection steps for replacing contact lenses / changing facial makeup are as follows: Get the sequence of user operations within the most recent preset time period, such as zooming in on the target interface to view virtual contact lenses and virtual facial makeup. The operation frequency of the parsing operation record sequence; When the operation frequency does not exceed the frequency threshold, a first quick selection list is generated based on the first replacement material library of virtual contact lenses / virtual face makeup that the user does not like to try on, and the user can quickly select replacement contact lenses / replace face makeup from the first quick selection list; Otherwise, the filter time period of the parsing operation record sequence is used; where the first and last times of the filter time period are the two times with the greatest distance between them among multiple times when the same area is viewed in a local magnified view for more than the dwell time threshold. Parse the set of operation elements in the operation region of the operation record sequence within the filtered time period; Based on the set of operation elements, match the second replacement material library; Based on the second replacement material library, a second quick selection list is generated, allowing users to quickly select from the second quick selection list to replace contact lenses or facial makeup.
5. A method for collaborative recommendation of virtual try-on for contact lenses and intelligent makeup, characterized in that, include: Obtain the user's facial features and matching preferences; Based on facial features and matching needs, we recommend matching contact lenses and facial makeup for users; It assists users in virtually trying on contact lenses and virtually applying facial makeup. It assists users in virtually trying on contact lenses and virtually applying facial makeup, including: Separately plan the virtual application rhythm of facial makeup and the virtual timing of wearing contact lenses; Based on the virtual face-applying rhythm, the interactive system assists users in applying facial makeup. During the interactive process of assisting the user in applying facial makeup, when the virtual wearing opportunity arrives, the interactive user virtually tries on contact lenses. The separately planned virtual application rhythm of facial makeup and virtual wearing timing of contact lenses include: Analyzing multiple makeup elements of facial makeup; Based on the constraints of makeup element matching, various makeup elements are matched to obtain multiple makeup slices; Based on the serialization processing constraints, each makeup slice is serialized to obtain a makeup slice sequence; By imbuing the makeup slice sequence with a rhythm control mechanism, a virtual face-applying rhythm can be obtained; The virtual wearing time for contact lenses is generated when the user's average actual acceptance of the virtual face-wearing of makeup slices for a consecutive threshold number of times exceeds the actual acceptance threshold. The constraints on the combination of makeup elements include: The same makeup element is contained in at least two makeup slices; In addition, each makeup slice must contain at least one makeup element related to the eyes; In addition, the similarity between any two makeup slices does not exceed the similarity threshold; The serialization processing constraints include: The user prediction preference and sequence order of each makeup slice in the makeup slice sequence are mapped into the curve coordinate system to obtain a complete peak; The rhythm control mechanism includes: Step 1: Apply each makeup slice in the makeup slice sequence to the face in the order they appear in the sequence; Step 2: When the i-th makeup slice in the makeup slice sequence is virtually applied to the face, the value of j is determined based on the user's actual acceptance level at that time; Step 3: Virtually apply the (i+j)th makeup slice in the makeup slice sequence to the face, and then use the (i+j)th makeup slice as the new ith makeup slice to return and repeat Step 2. This process is repeated. Where 1≤i+j≤N, and N is the total number of makeup slices in the makeup slice sequence.
6. The method for virtual try-on of contact lenses and intelligent makeup collaborative recommendation as described in claim 5, characterized in that, The facial features include at least: face shape, eyebrow shape, eye size and shape, distance between eyes, nose shape, lip shape, skin color, and facial proportions; The matching requirements include at least: travel scenarios, preferred makeup styles, types of activities participated in, and functional requirements for contact lenses.
7. The method for virtual try-on of contact lenses and intelligent makeup collaborative recommendation as described in claim 6, characterized in that, The system recommends matching contact lenses and facial makeup to users based on facial features and styling needs, including: Based on a pre-trained collaborative matching model, the system recommends matching contact lenses and facial makeup to users according to their facial features and matching needs. The pre-training steps for the collaborative matching model are as follows: A large amount of matching information labeled with standard matching schemes is used as training samples for machine learning training to obtain a synergistic matching model.
Citation Information
Patent Citations
Daily makeup design method and system based on face feature region recognition
CN102708575A
APP (application) interface display method and intelligent terminal
CN108228306A
Indoor design achievement evaluation method and system based on virtual reality technology
CN118587344A
Face recognition-based beauty makeup recommendation method and device, equipment and medium
CN119397109A
Systems and methods for simulated application of cosmetic effects
US20170255478A1