SYSTEM AND METHOD FOR FASHION ASSESSMENT AND RECOMMENDATION

Abstract

The present disclosure provides a fashion assessment and recommendation system (100) including an enclosed structure (102) with an open front designed to accommodate a user and an LED display (104) on the back panel that serves as both a mirror and a feedback display, along with multiple cameras (106) positioned on the top of the LED display (104) and side panels to capture images of the user from various angles. Speakers (108) are mounted on the side panels for audio feedback based on the system's (100) analysis. A microcontroller connects the LED display (104), cameras (106) and speakers (108) to manage the system’s (100) operations. The system (100) processes the captured images to analyze a set of first features and evaluate the user's current outfit by assessing a set of second features. Based on this analysis, system (100) generates personalized clothing recommendations, which are displayed on the LED screen, while audio feedback about the assessment and suggestions is delivered through the speakers (108).

Specification

Description:TECHNICAL FIELD
[0001] The present disclosure relates, in general, to the field of image processing and artificial intelligence. More particularly, it relates to system and method fashion assessment and recommendation that utilize image processing, machine learning algorithms, and user feedback mechanisms to analyze an individual's attire and provide personalized clothing suggestions based on user-specific characteristics and preferences.

BACKGROUND
[0002] The following description of the related art is intended to provide background information pertaining to the field of the disclosure. This section may include certain aspects of the art that may be related to various features of the present disclosure. However, it should be appreciated that this section be used only to enhance the understanding of the reader with respect to the present disclosure, and not as admission of the prior art.
[0003] The intersection of fashion and technology has rapidly evolved in recent years, with machine learning and computer vision making significant strides in transforming the fashion industry. Technologies that analyse visual data have enabled the development of recommendation systems that assist users in selecting appropriate clothing styles based on personal features such as body shape, skin tone, and style preferences. Traditional fashion recommendation systems typically focus on a single aspect of the user's appearance, such as color coordination or body measurements, without considering the holistic view of the individual's appearance and style preferences. While these systems provide some level of assistance, they often lack the ability to deliver comprehensive and personalized recommendations that account for the complexities of individual features.
[0004] Convolutional Neural Networks (CNNs) have become an essential tool in image classification tasks, including applications like skin tone detection and body shape analysis. Existing systems that rely on CNNs are capable of identifying basic visual patterns, but their scope is generally limited to individual aspects of a user's appearance, such as skin color or clothing type. Additionally, many of these systems struggle with handling the subjective nuances of beauty, style, and personal preference, which are vital when recommending fashion choices. They often lack the ability to make recommendations that consider how different aspects of an individual's appearance, such as skin tone, body shape, and facial features, interact with one another to influence suitable fashion choices.
[0005] One of the main disadvantages of existing fashion recommendation systems is their lack of adaptability and personalization. These systems generally operate using predefined models that do not adapt to the user's changing preferences or the nuances in body features. Additionally, many current systems offer static recommendations that do not improve over time based on feedback from users. This limitation leads to generalized recommendations that may not be accurate or tailored to an individual's unique attributes. Furthermore, such systems often ignore key factors such as real-time environmental conditions (e.g., lighting and humidity) that can impact the accuracy of image analysis, resulting in suboptimal fashion advice.
[0006] One of the main disadvantages of existing fashion recommendation systems is their lack of adaptability and personalization. These systems generally operate using predefined models that do not adapt to the user's changing preferences or the nuances in body features. Additionally, many current systems offer static recommendations that do not improve over time based on feedback from users. This limitation leads to generalized recommendations that may not be accurate or tailored to an individual's unique attributes. Furthermore, such systems often ignore key factors such as real-time environmental conditions (e.g., lighting and humidity) that can impact the accuracy of image analysis, resulting in suboptimal fashion advice.
[0007] There is, therefore, a need to overcome the above-mentioned drawbacks, shortcomings, and limitations associated with existing systems by providing a simple, reliable, and effective solution for a multi-faceted analysis by considering body shape, skin tone, face shape, and the type of clothing worn and learn from user feedback and iteratively improve recommendations.

OBJECTS OF THE PRESENT DISCLOSURE
[0008] A general object of the present disclosure is to provide personalized clothing recommendations to users by analyzing individual features such as skin tone, body shape, and outfit coordination.
[0009] An object of the present disclosure is to integrate advanced image processing and artificial intelligence techniques, enabling accurate multi-angle image capture and feature analysis for providing tailored fashion advice.
[0010] An object of the present disclosure is to provide real-time visual and audio feedback to help users to understand the suitability of their present outfit and guiding them toward better fashion choices.
[0011] An object of the present disclosure is to incorporate environmental and personal data into the fashion recommendation process, ensuring that clothing suggestions are appropriate for the user's body type and current weather or atmospheric conditions.

SUMMARY
[0012] Aspects of the present disclosure relates to the technical field of image processing and artificial intelligence. More particularly, it relates to system and method fashion assessment and recommendation that utilize image processing, machine learning algorithms, and user feedback mechanisms to analyze an individual's attire and provide personalized clothing suggestions based on user-specific characteristics and preferences.
[0013] According to an aspect, a fashion assessment and recommendation system is disclosed. The system includes an enclosed structure with an open front configured to accommodate a user, an LED display located on the back panel of the enclosure and configured to function as a mirror for the user and to display visual feedback, a set of speakers positioned on side panels of the enclosure to provide audio feedback to the user based on system's assessment of the user's outfit and one or more cameras positioned on the top of the LED display and the side panels, configured to capture images of the user from different angle. In addition, the system includes a microcontroller communicatively coupled to the LED display, the set of speakers and the one or more cameras, where the microcontroller is configured to receive and process an image captured by the cameras to analyze a set of first features and evaluate the user's present outfit by assessing a set of second features using the captured image data. Further, the microcontroller generates personalized clothing recommendations based on the set of first and second features, displays the recommendations and results on the LED display and outputs audio feedback through the speakers or headset regarding the analysis and recommendations.
[0014] The one or more cameras may include a first camera positioned at the top of the LED display to capture a frontal image of the user and a second and third camera positioned on the side panels to capture side images of the user from different angles.
[0015] The system may include a weighing machine integrated into the floor of the enclosure, the weighing machine communicatively coupled to the microcontroller to provide weight data as part of the user's fashion assessment.
[0016] The set of first features may include skin tone and body shape, and the set of second features may include color coordination, fit, and style relevance of the user's present outfit.
[0017] The system may include a database of fashion styles and clothing options, wherein the microcontroller is configured to generate the personalized clothing recommendations by comparing the user's analyzed features to entries in the database.
[0018] The system may include a proximity sensor configured to detect the position and movement of the user within the enclosure, the proximity sensors enable automatic activation of the cameras and system functions when the user enters the enclosure.
[0019] The system may include a temperature sensor integrated into the enclosure, the temperature data being used by the microcontroller to recommend clothing suitable for the detected environmental conditions.
[0020] The system may include a barometer integrated into the enclosure, the barometer configured to measure atmospheric pressure, and the microcontroller configured to use the atmospheric pressure data to recommend weather-appropriate clothing for the use.
[0021] In an another aspect, a method for providing fashion assessment and recommendations is disclosed. The method includes receiving an image of a user captured by one or more cameras positioned within an enclosed structure having an open front, processing the captured image to analyze a set of first features, where the first features relate to user characteristics including but not limited to body shape, skin tone, and facial features, and evaluating the user's current outfit by assessing a set of second features from the captured image, where the second features include color, style, and fit of the clothing worn by the use. In addition, the method includes generating personalized clothing recommendations based on the analysis of the first features and the evaluation of the second features and displaying the personalized clothing recommendations on an LED display located on a back panel of the enclosure, which is configured to function as a mirror for the user. Further, the method includes providing audio feedback through a set of speakers positioned on side panels of the enclosure, where the audio feedback is based on the analysis of the user's outfit and the generated recommendations and outputting visual feedback related to the fashion assessment on the LED display to enhance user experience.
[0022] Various objects, features, aspects and advantages of the inventive subject matter will become more apparent from the following detailed description of preferred embodiments, along with the accompanying drawing figures in which like numerals represent like components.

BRIEF DESCRIPTION OF THE DRAWINGS
[0023] The accompanying drawings are included to provide a further understanding of the present disclosure and are incorporated in and constitute a part of this specification. The drawings illustrate exemplary embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure.
[0024] FIGs. 1A-1C illustrate exemplary views of the proposed fashion assessment and recommendation system, in accordance with an embodiment of the present disclosure.
[0025] FIG. 2 illustrates an exemplary block diagram of the proposed fashion assessment and recommendation system, in accordance with an embodiment of the present disclosure.
[0026] FIG. 3 illustrates an exemplary work flow chart of the fashion assessment and recommendation system, in accordance with an embodiment of the present disclosure.
[0027] FIG. 4 illustrates an exemplary flow chart for implementing the steps of the proposed method for providing fashion assessment and recommendations, in accordance with an embodiment of the present disclosure.
[0028] FIG. 5 illustrates an exemplary flowchart demonstrating the functional integration of the proposed system, in accordance with an embodiment of the present disclosure.

DETAILED DESCRIPTION
[0029] The following is a detailed description of embodiments of the disclosure depicted in the accompanying drawings. The embodiments are in such details as to clearly communicate the disclosure. However, the amount of detail offered is not intended to limit the anticipated variations of embodiments; on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the present disclosure as defined by the appended claims.
[0030] Embodiments of the present disclosure explained herein relate to the field of image processing and artificial intelligence. More particularly, it relates to system and method fashion assessment and recommendation that utilize image processing, machine learning algorithms, and user feedback mechanisms to analyze an individual's attire and provide personalized clothing suggestions based on user-specific characteristics and preferences.
[0031] Referring to FIGs. 1A-1C and FIG. 2, the proposed fashion assessment and recommendation system 100 is disclosed. The system 100 can include an enclosed structure 102 with an open front configured to accommodate a user, an LED display 104 located on the back panel of the enclosure and configured to function as a mirror for the user and to display visual feedback, a set of speakers 108 positioned on side panels of the enclosure to provide audio feedback to the user based on system's 100 assessment of the user's outfit and one or more cameras 106 positioned on the top of the LED display 104 and the side panels, configured to capture images of the user from different angle. In addition, the system 100 can include a microcontroller communicatively coupled to the LED display 104, the set of speakers 108 and the one or more cameras 106, where the microcontroller can be configured to receive and process an image captured by the cameras 106 to analyze a set of first features and evaluate the user's present outfit by assessing a set of second features using the captured image data. Further, the microcontroller can generate personalized clothing recommendations based on the set of first and second features, display the recommendations and results on the LED display 104 and output audio feedback through the speakers 108 regarding the analysis and recommendations.
[0032] In an embodiment, the one or more cameras 106 can include a first camera positioned at the top of the LED display 104 to capture a frontal image of the user and a second and third camera positioned on the side panels to capture side images of the user from different angles.
[0033] In an embodiment, the system 100 can include a weighing machine integrated into the floor of the enclosure, the weighing machine communicatively coupled to the microcontroller to provide weight data as part of the user's fashion assessment.
[0034] In an embodiment, the set of first features can include skin tone and body shape, and the set of second features can include color coordination, fit, and style relevance of the user's present outfit.
[0035] In an embodiment, the system 100 can include a database of fashion styles and clothing options, where the microcontroller is configured to generate the personalized clothing recommendations by comparing the user's analyzed features to entries in the database.
[0036] In an embodiment, the system 100 can include a proximity sensor configured to detect the position and movement of the user within the enclosure, the proximity sensors enable automatic activation of the cameras 106 and system 100 functions when the user enters the enclosure.
[0037] In an embodiment, the system 100 can include a temperature sensor integrated into the enclosure, the temperature data being used by the microcontroller to recommend clothing suitable for the detected environmental conditions.
[0038] In an embodiment, the system 100 can include a barometer integrated into the enclosure, the barometer configured to measure atmospheric pressure, and the microcontroller configured to use the atmospheric pressure data to recommend weather-appropriate clothing for the use.
[0039] In an embodiment, the system 100 can include a humidity sensor configured to measure the ambient humidity within the environment. The microcontroller can be configured to use the humidity data to adjust system 100 operations or recommend clothing suitable for the detected humidity level.
[0040] In an embodiment, the system 100 can include an infrared (IR) sensor configured to receive infrared signals from an external entertainment device. The microcontroller can be configured to process these signals to control the entertainment device by generating and transmitting corresponding infrared output signals through an IR LED.
[0041] In an embodiment, the system 100 can include a LiDAR sensor configured to detect and map the user's surroundings in three dimensions. The microcontroller can use the LiDAR data to enhance image processing and provide more accurate clothing fit recommendations based on the user's body measurements and spatial positioning.
[0042] In an embodiment, the system 100 can include a Bluetooth headset communicatively coupled to the microcontroller. The microcontroller can be configured to transmit audio feedback and recommendations to the user through the Bluetooth headset, enabling hands-free interaction with the system 100.
[0043] In an embodiment, the system 100 can include a LiDAR sensor configured to emit laser pulses to measure distances between the user and the surrounding objects. The microcontroller can process the LiDAR data to create a three-dimensional model of the user's body and environment.
[0044] In an embodiment, the system 100 can include a HDMI and USB ports, wherein the HDMI port is configured to connect the system 100 to external displays for enhanced visual output, and the USB port can be configured to allow external devices to transfer data to or from the system 100, enabling additional functionality such as software updates or media integration.
[0045] In an embodiment, the system 100 can include LED lights positioned around the camera modules. The LED lights can be configured to automatically adjust their brightness based on ambient lighting conditions detected by the proximity sensor, to provide optimal illumination for image capture by the camera modules.
[0046] Referring to FIG. 3, work flow chart 300 of the fashion assessment and recommendation system is disclosed. The system can start with capturing images of the user using one or more cameras. The captured images can undergo processing to extract relevant features, such as the user's body shape and skin tone. A Convolutional Neural Network (CNN) model can be employed to analyze the user's skin tone. The model can process the image to determine the specific skin tone features. A second CNN model can be used for identifying the user's body shape. This model can help in recognizing the body structure based on the input images. Relevant features from both skin tone analysis and body shape determination can be extracted. These features can be critical for determining the user's style preferences and for generating outfit recommendations. Using the extracted features, the system can analyze the clothing style most appropriate for the user, based on body shape and skin tone. The system can evaluate the user's current outfit, assigning a rating based on how well it aligns with the determined style preferences. The rating data, along with other extracted features, can be stored in an Excel sheet for record-keeping or further analysis. The system can also utilize a Recurrent Neural Network (RNN) model for advanced outfit recommendations based on past data. Based on the rating, style determination, and previous preferences (as processed by the RNN), the system can generate personalized outfit suggestions for the user. After analysing all the data and feedback, the system can provides suggestions for a better outfit to the user.
[0047] Referring to FIG. 4, method 400 for providing fashion assessment and recommendations is disclosed. Method can be performed by the components of the system 100 of FIGs. 1A-1C and FIG. 2. Method 400 can include step 402 of receiving an image of a user captured by one or more cameras positioned within an enclosed structure having an open front. In addition, the method 400 can include step 404 of processing the captured image to analyze a set of first features. The first features can relate to user characteristics including but not limited to body shape, skin tone, and facial features. Further, the method 400 can include step 406 of evaluating the user's current outfit by assessing a set of second features from the captured image. The second features can include color, style, and fit of the clothing worn by the use.
[0048] In an embodiment, at step 408 the method 400 can include evaluating the user's current outfit by assessing a set of second features from the captured image. The second features can include color, style, and fit of the clothing worn by the use. The method 400 can further include step 410 of generating personalized clothing recommendations based on the analysis of the first features and the evaluation of the second features. In addition, the method 400 can include step 412 of displaying the personalized clothing recommendations on an LED display located on a back panel of the enclosure. The display can be configured to function as a mirror for the user. At step 414 the method 400 can include providing audio feedback through a set of speakers positioned on side panels of the enclosure. The audio feedback can be based on the analysis of the user's outfit and the generated recommendations. Further, the method 400 can include step 416 of outputting visual feedback related to the fashion assessment on the LED display to enhance user experience.
[0049] Referring to FIG. 5, flowchart 500 demonstrating the functional integration of the proposed system is disclosed. The system can start by capturing an image of the user through a camera. This image can be used as the primary input for further processing and analysis. Once the image is captured, it can undergo initial processing, which can involve standard image processing techniques like resizing, adjusting contrast, etc. Next, any noise present in the image (irrelevant data or visual artifacts) can be reduced to ensure better analysis in subsequent steps. The image can be segmented into various regions, focusing on the body and skin to enable specific feature extraction, which can help in determining the user's body shape and skin tone. Further, the system checks whether skin tone analysis is necessary. If not required, the flow skips to body shape analysis, where if skin tone analysis can be required, the system can use a convolutional neural network (CNN Model 2) to extract relevant skin tone features. If skin tone analysis is not needed, the process can directly proceed to body shape analysis.
[0050] The system can check if body shape analysis is needed. If required, it can proceed with the analysis, otherwise skip to feature extraction, where the system can use CNN Model 1 to extract body shape features. If body shape analysis is unnecessary, it can directly moves to the next step. Next, the system can extract skin tone features if skin tone analysis is carried out and similarly, body shape features can be extracted if body shape analysis is performed. The system can combine the extracted features to generate a comprehensive user profile and combined features can be cross-referenced with a style database that stores information on various clothing styles, colors, and patterns suited for different body shapes and skin tones. Further, the system can check whether the style database is current, where if the database is up-to-date, it proceeds with style determination and if the database is outdated, it can trigger a database update process. Using the cross-referenced data, the system can determine the optimal clothing styles that suit the user.
[0051] The system can finalize the best clothing style based on the analysis and features extracted. The system can check whether the user's current outfit matches the style determined by the system, where if the outfit can match the optimal style, it can be assigned a high rating and if the outfit does not match, the system evaluates mismatches and suggests improvements. Next, if the outfit can match the determined style, a high rating can be assigned and if the outfit doesn't match, the system can log the mismatches and can suggest improvements based on the style database. A detailed outfit report can be generated, summarizing the analysis and recommendations. Further, the report can be sent to the user for review and the system can log feedback and the satisfaction score provided by the user.
[0052] The system can evaluate whether the feedback is positive, where if feedback is positive, it can store the successful analysis parameters and if feedback is negative, the system can trigger a Recurrent Neural Network (RNN) model for adaptive learning to improve its recommendations. Next, if the feedback was negative, the system can retrain itself using the feedback to better adapt to the user's preferences. In cases of positive feedback, the system can store the successful analysis parameters for future recommendations. Based on the feedback, the system can update the user's profile with new preferences, ensuring more accurate recommendations in the future. Further, the process can conclude once the recommendations is made and feedback is logged, with the option for adaptive learning triggered if necessary.
[0053] Thus, the present disclosure provides a system and method for fashion assessment and recommendation which captures user images and performs detailed analyses of body shape and skin tone through CNN models, and tailors outfit suggestions to individual preferences. The use of an up-to-date style database ensures relevance, while feedback loops allow for continuous improvement. With adaptive learning powered by an RNN model, the present disclosure evolves based on user feedback, offering increasingly accurate recommendations over time.
[0054] While the foregoing describes various embodiments of the invention, other and further embodiments of the invention may be devised without departing from the basic scope thereof. The scope of the invention is determined by the claims that follow. The invention is not limited to the described embodiments, versions or examples, which are included to enable a person having ordinary skill in the art to make and use the invention when combined with information and knowledge available to the person having ordinary skill in the art.

ADVANTAGES FOR THE PRESENT DISCLOSURE
[0055] The present disclosure provides utilization of multiple cameras placed at different angles to capture the user's image, ensuring a thorough analysis of their outfit from all sides, leading to more accurate feedback.
[0056] The present disclosure provides assessment of features such as body shape, skin tone, color coordination, and style relevance, the system provides customized clothing suggestions tailored to the user's specific attributes.
[0057] The present disclosure utilizes a LED display functioning as both a mirror and a feedback screen, allowing the user to visually see the recommendations and assessments in real-time.
[0058] The present disclosure enables recommending clothing based on current environmental condition.
[0059] The present disclosure enables providing additional data for more accurate outfit recommendations by factoring in the user's body profile and ensuring better clothing fit advice.
, Claims:1. A fashion assessment and recommendation system (100), comprising:
an enclosed structure (102) with an open front configured to accommodate a user;
an LED display (104) located on the back panel of the enclosure, the LED display (104) configured to function as a mirror for the user and to display visual feedback;
one or more cameras (106) positioned on the top of the LED display (104) and the side panels, configured to capture images of the user from different angles;
a set of speakers (108) positioned on side panels of the enclosure to provide audio feedback to the user based on system's (100) assessment of the user's outfit; and
a microcontroller communicatively coupled to the LED display (104), the one or more cameras (106), and the set of speakers (108), the microcontroller configured to:
receive and process an image captured by the cameras (106) to analyze a set of first features;
evaluate the user's present outfit by assessing a set of second features using the captured image data;
generate personalized clothing recommendations based on the set of first and second features;
display the recommendations and results on the LED display (104); and
output audio feedback through the speakers (108) regarding the analysis and recommendations.

2. The system (100) as claimed in claim 1, wherein the one or more cameras (106) comprise:
a first camera positioned at the top of the LED display (104) to capture a frontal image of the user; and
a second and third camera positioned on the side panels to capture side images of the user from different angles.

3. The system (100) as claimed in claim 1, wherein the system (100) comprises a weighing machine integrated into the floor of the enclosure, the weighing machine communicatively coupled to the microcontroller to provide weight data as part of the user's fashion assessment.

4. The system (100) as claimed in claim 1, wherein the set of first features comprises skin tone and body shape, and the set of second features comprises color coordination, fit, and style relevance of the user's present outfit.

5. The system (100) as claimed in claim 1, wherein the system (100) comprises a database of fashion styles and clothing options, wherein the microcontroller is configured to generate the personalized clothing recommendations by comparing the user's analyzed features to entries in the database.

6. The system (100) as claimed in claim 1, wherein the system (100) comprises a proximity sensor configured to detect the position and movement of the user within the enclosure, the proximity sensors enable automatic activation of the cameras (106) and system (100) functions when the user enters the enclosure.

7. The system (100) as claimed in claim 1, wherein the system (100) comprises a temperature sensor integrated into the enclosure, the temperature data being used by the microcontroller to recommend clothing suitable for the detected environmental conditions.

8. The system (100) as claimed in claim 1, wherein the system (100) comprises a barometer integrated into the enclosure, the barometer configured to measure atmospheric pressure, and the microcontroller configured to use the atmospheric pressure data to recommend weather-appropriate clothing for the use.

9. A method (400) for providing fashion assessment and recommendations, comprising:
receiving (402) an image of a user captured by one or more cameras positioned within an enclosed structure having an open front;
processing (404) the captured image to analyze a set of first features, wherein the first features relate to user characteristics including but not limited to body shape, skin tone, and facial features;
evaluating (406) the user's current outfit by assessing a set of second features from the captured image, wherein the second features include color, style, and fit of the clothing worn by the use;
generating (408) personalized clothing recommendations based on the analysis of the first features and the evaluation of the second features;
displaying (410) the personalized clothing recommendations on an LED display located on a back panel of the enclosure, which is configured to function as a mirror for the user;
providing (412) audio feedback through a set of speakers positioned on side panels of the enclosure, wherein the audio feedback is based on the analysis of the user's outfit and the generated recommendations; and
outputting (414) visual feedback related to the fashion assessment on the LED display to enhance user experience.

Documents