Method and System for Recommending Dog Clothing Using Review Data-Based Dog Clothing Size Reference Information

Abstract

The present invention relates to a method for recommending dog clothing using standard size information for dog clothing based on review data. More specifically, it relates to a method for recommending the most suitable clothing to new buyers by resolving size deviations or labeling errors by automatically correcting clothing size information provided by a seller based on the actual body measurements of a dog derived by analyzing images and dog information from the review data of buyers who have purchased clothing products.

Description

Technology Field

[0001] The present invention relates to a method for recommending dog clothing using standard size information for dog clothing based on review data. More specifically, it relates to a method for recommending the most suitable clothing to new buyers by resolving size deviations or labeling errors by automatically correcting clothing size information provided by a seller based on the actual body measurements of a dog derived by analyzing images and dog information from the review data of buyers who have purchased clothing products. Background Technology

[0003] With the recent increase in the perception of pets as members of the family, the pet supplies market, particularly the pet clothing market, is growing rapidly. Unlike the past practice of trying on items in offline stores before purchasing, online commerce has recently become the dominant method for purchasing clothing, offering a variety of designs and competitive pricing.

[0004] However, purchasing dog clothing online faces the problem of a very high rate of returns and exchanges due to incorrect size selection. This is because, unlike human clothing, a dog's body structure varies greatly depending on breed, body type, and fur length, and the standards for size charts (e.g., S, M, L, etc.) provided by different brands or manufacturers are not standardized.

[0005] Consequently, many consumers refer to review data left by other buyers to make purchasing decisions. However, existing review systems rely on unstructured text data, such as "It fits a 5kg poodle well" or "It seems to run a bit small," or simple photos of the item being worn, making it difficult to utilize this as quantitative data. In other words, since it is impossible to know whether the reviewer's dog has a standard body type, is obese, or has a thick coat, the reliability of the information is low, making it difficult for new buyers to directly apply it to their own dogs for judgment.

[0006] Conventional technologies primarily relied on a method where users manually measured their dog's body dimensions (neck circumference, chest circumference, and back length) using a tape measure and inputted them, which were then simply compared to a size chart posted by the seller to recommend the closest size. Alternatively, technologies utilizing artificial intelligence have recently been proposed to automatically measure body dimensions through image analysis when a photo of the dog is taken.

[0007] However, even if these conventional technologies measure the user's dog's dimensions with great precision, if the seller's clothing size information used for comparison is inaccurate or does not reflect the dog's actual fit, the user will be recommended an incorrect size even if they input their dog's exact dimensions.

[0008] Therefore, there is an urgent need to develop an advanced clothing recommendation system that does not simply rely on the seller's input information, but instead reverse-analyzes review data from numerous buyers who have actually purchased and worn the clothing to correct the seller's size information and even considers factors such as fur compression rates and skeletal characteristics specific to each breed. Prior art literature

[0010] Korean Patent Publication KR 10-2025-0076995 A “Method and device for providing clothing size information for companion dogs” (May 30, 2025) The problem to be solved

[0011] The present invention relates to a method for recommending dog clothing using standard size information for dog clothing based on review data. More specifically, the invention aims to provide a method for recommending dog clothing using standard size information for dog clothing based on review data, which recommends the most suitable clothing to new buyers by fundamentally resolving size deviations or labeling errors by analyzing images and dog information from the review data of buyers who have purchased clothing products, and by automatically correcting the clothing size information provided by the seller based on the actual body measurements of the dog derived from analyzing the dog's actual body measurements. means of solving the problem

[0013] To solve the above problem, a method for recommending dog clothing based on review data analysis is performed on a server system comprising one or more processors and one or more memories, wherein the server system includes a database storing actual total body lengths stored according to a combination of clothing information, including body circumference, body length, and neck circumference for each of a plurality of clothing products received from a seller, and a plurality of dog breed information, weight information, and body shape information; and the dog clothing recommendation method comprises: a review data collection step of collecting a plurality of review data including one or more of dog breed information, weight information, and image information of a dog from a buyer who has purchased the clothing products; and an actual length derivation step of deriving the total body length, body circumference, body length, and neck circumference identified in an image from the image information included in each of the plurality of review data, and deriving the actual body circumference, actual body length, and actual neck circumference of each buyer's dog based on the total body length and a plurality of actual total body lengths stored in the database. A method for recommending clothing for a pet dog is provided, comprising: a clothing information update step for deriving updated clothing information based on the actual body circumference, actual body length, and actual neck circumference for each of the plurality of review data, wherein the clothing information entered by the seller for the clothing product is updated; and a clothing recommendation step for recommending the clothing product to the new buyer by comparing the actual body circumference, actual body length, and actual neck circumference for the new buyer's pet dog derived based on input information entered by the new buyer with the updated clothing information.

[0014] In one embodiment of the present invention, the clothing information update step may include: a statistical calculation step for calculating the average and standard deviation for each of the actual torso circumference, actual torso length, and actual neck circumference of the plurality of review data derived in the actual length derivation step; a range calculation step for calculating a range interval equal to a preset multiple of the standard deviation centered on the average for each of the actual torso circumference, actual torso length, and actual neck circumference, thereby deriving a first range interval for the actual torso circumference, a second range interval for the actual torso length, and a third range interval for the actual neck circumference; and an update information generation step for generating updated clothing information that updates the clothing information entered by the seller based on the first range interval, the second range interval, and the third range interval.

[0015] In one embodiment of the present invention, the actual length derivation step comprises: a length calculation step in which image information included in each of the plurality of review data is input into a pre-trained deep learning model to derive values ​​corresponding to torso circumference, torso length, neck circumference, and total body length based on a plurality of feature points, thereby deriving torso circumference, torso length, neck circumference, and total body length corresponding to each of the plurality of image information; and a ratio calculation step in which, for each of the torso circumference, torso length, neck circumference, and total body length corresponding to each of the plurality of image information, a first ratio information regarding the ratio of torso circumference to total body length, a second ratio information regarding the ratio of torso length to total body length, and a third ratio information regarding the ratio of neck circumference to total body length are calculated. and may include an actual length calculation step of deriving a matching total body length that matches each of the plurality of review data among the actual total body lengths included in the database, and for each of the plurality of review data, calculating the actual torso circumference, actual torso length, and actual neck circumference of each dog in the review data based on the matching total body length, the first ratio information, the second ratio information, and the third ratio information.

[0016] In one embodiment of the present invention, the total body length corresponds to the length from a first feature point referring to the tip of the dog's head to a second feature point referring to the starting point of the tail, the body circumference corresponds to the circumference at a third feature point corresponding to the thickest part of the dog's chest, the body length corresponds to the length measured along the back line from a fourth feature point referring to the starting point where the dog's neck and back meet to a fifth feature point referring to the starting point of the tail, and the neck circumference corresponds to the circumference measured based on a sixth feature point corresponding to the lowest part of the neck where the dog's neck and shoulder are connected.

[0017] In one embodiment of the present invention, the input information includes one or more of the breed information, weight information, and image information of the new buyer's dog, and the clothing recommendation step may further include: a step of extracting a first layer corresponding to an external silhouette including the dog's fur and a second layer corresponding to an internal silhouette excluding the dog's fur from the image information included in the input information; and a step of calculating a fur length compressible by the dog's fur based on the distance difference between the first layer and the second layer.

[0018] In one embodiment of the present invention, the clothing recommendation step may include: a step of calculating the actual torso circumference, actual torso length, and actual neck circumference based on the input information; a step of calculating the torso circumference excluding hair and the neck circumference excluding hair by subtracting the hair length from the actual torso circumference and the actual neck circumference, respectively, when the hair length exceeds a preset hair length threshold; and a step of providing the corresponding clothing product when the torso circumference excluding hair, the neck circumference excluding hair, and the actual torso length correspond to the range included in the updated clothing information, by comparing the torso circumference excluding hair, the neck circumference excluding hair, and the actual torso length.

[0019] In one embodiment of the present invention, the input information includes one or more of the breed information, weight information, and image information of the new buyer's dog, and the clothing recommendation step may include: a step of deriving the distance between the eyes corresponding to a part with minimal change in skeletal size according to the body shape information and the dog's breed from the image information included in the input information when the breed information included in the input information corresponds to a mixed dog with two or more breeds; a step of deriving the ratio of the distance between the eyes to the ratio of the total body length; and a body length correction step of calculating a correction value based on the ratio of the distance between the eyes to the average value of a plurality of actual total body lengths derived based only on the weight information and body shape information included in the input information in the database, and calculating the actual torso circumference, actual torso length, and actual neck circumference of the mixed dog, respectively, based on the correction value.

[0020] To solve the above problem, a review data analysis-based dog clothing recommendation system comprising one or more processors and one or more memories comprises: a database storing actual total body lengths stored according to a combination of clothing information, including body circumference, body length, and neck circumference for each of a plurality of clothing products received from a seller, and a plurality of dog breed information, weight information, and body shape information; a review data collection unit collecting a plurality of review data including one or more of dog breed information, weight information, and image information of a dog from a buyer who has purchased the clothing products; and an actual length derivation unit deriving the total body length, body circumference, body length, and neck circumference identified in an image from the image information included in each of the plurality of review data, and deriving the actual body circumference, actual body length, and actual neck circumference of each buyer's dog based on the total body length and a plurality of actual total body lengths stored in the database. A pet dog clothing recommendation system comprises: a clothing information update unit that derives updated clothing information based on the actual body circumference, actual body length, and actual neck circumference of the plurality of review data, based on clothing information entered by the seller for the above clothing product; and a clothing recommendation unit that recommends the above clothing product to the new buyer by comparing the actual body circumference, actual body length, and actual neck circumference of the new buyer's pet dog derived based on input information entered by the new buyer with the updated clothing information. Effects of the invention

[0022] According to one embodiment of the present invention, by updating clothing information based on actual body measurements obtained by reverse-calculating review data from multiple users who actually purchased the product, rather than relying on unilateral clothing size information provided by the seller, it is possible to achieve the effect of establishing realistic fit standards that reflect the manufacturer's measurement errors or material characteristics.

[0023] According to one embodiment of the present invention, when measuring the body dimensions of a pet dog, the length is derived based on clear anatomical feature points such as the head, tail, and back line, thereby eliminating errors caused by the user and achieving the effect of securing consistent dimension data.

[0024] According to one embodiment of the present invention, by setting a statistical interval using the mean and standard deviation of dimension information derived from review data and performing a preprocessing process to remove outliers, it is possible to prevent information distortion caused by extreme review data and achieve the effect of suggesting a size range suitable for the majority of users.

[0025] According to one embodiment of the present invention, when analyzing review images, a deep learning model automatically extracts actual dimensions based on ratios by linking feature points within the image with a database without the need for a separate measurement tool, thereby enabling the effect of converting a vast amount of review data into quantitative size data and accumulating it into a systematic database without requiring the reviewer to input separate dimensions.

[0026] According to one embodiment of the present invention, by performing a process of calculating hair length by distinguishing between the external silhouette and the internal silhouette when analyzing an image of a new buyer's pet dog, it is possible to prevent errors in which the actual skin dimensions are inflated in the case of breeds with thick fur and to provide a precise size recommendation that takes into account compressibility when wearing clothing.

[0027] According to one embodiment of the present invention, in the case of mixed-breed dogs for which standardized data is lacking, by applying a logic that corrects the total body length based on the distance between the eyes, which has minimal change in skeletal size, it is possible to provide a recommendation service of high reliability that is universally applicable to all dog breeds.

[0028] According to one embodiment of the present invention, a seller can achieve the effect of lowering the return rate and increasing customer satisfaction through size information that is automatically optimized as reviews accumulate, without having to go through a separate size modification process. Brief explanation of the drawing

[0030] FIG. 1 illustrates a server system that performs a method for recommending dog clothing based on review data analysis according to an embodiment of the present invention. FIG. 2 illustrates the steps of performing a method for recommending dog clothing based on review data analysis according to an embodiment of the present invention. FIG. 3 illustrates a plurality of feature points and length definitions for deriving the total body length, torso circumference, torso length, and neck circumference according to one embodiment of the present invention. FIG. 4 illustrates an example of collecting a plurality of review data in a review data collection step according to an embodiment of the present invention. FIG. 5 illustrates an example of performing a length derivation step and a ratio calculation step from image information of review data according to an embodiment of the present invention. FIG. 6 illustrates an example of performing an actual length calculation step by referring to a database according to an embodiment of the present invention. FIG. 7 illustrates an example of performing a statistical calculation step and a segment calculation step based on the actual torso circumference, actual torso length, and actual neck circumference derived for each of the plurality of review data according to one embodiment of the present invention. FIG. 8 illustrates an example of generating updated clothing information by performing a clothing information update step according to an embodiment of the present invention. FIG. 9 illustrates an example of calculating the actual body circumference, actual body length, and actual neck circumference of a new buyer's dog based on input information of a new buyer according to an embodiment of the present invention. FIG. 10 illustrates an example of performing a clothing recommendation step by comparing size information and updated clothing information for a new buyer's pet dog according to one embodiment of the present invention. FIG. 11 illustrates an example of extracting a first layer and a second layer from image information and calculating hair length according to an embodiment of the present invention. FIG. 12 illustrates a process of calculating the body circumference excluding hair and the neck circumference excluding hair by comparing the hair length and the hair length threshold according to one embodiment of the present invention. FIG. 13 illustrates an example of deriving the distance between the eyes and the ratio of the two eyes when the input information according to one embodiment of the present invention corresponds to a mixed dog. FIG. 14 illustrates an example of calculating the total body length based on both eyes and the average total body length, respectively, to perform a body length correction step according to an embodiment of the present invention. FIG. 15 illustrates the process of calculating the final actual torso circumference, actual torso length, and actual neck circumference based on the difference value and correction threshold value according to one embodiment of the present invention. FIG. 16 schematically illustrates the internal configuration of a computing device according to one embodiment of the present invention. Specific details for implementing the invention

[0031] Hereinafter, various embodiments and / or aspects are disclosed with reference to the drawings. For illustrative purposes, numerous specific details are disclosed in the following description to aid in a general understanding of one or more aspects. However, it will also be recognized by those skilled in the art that these aspects may be practiced without such specific details. The following description and the accompanying drawings describe specific exemplary aspects of one or more aspects in detail. However, these aspects are exemplary, and some of the various methods in the principles of the various aspects may be used, and the description is intended to include all such aspects and their equivalents.

[0033] In addition, various aspects and features will be presented by a system that may include a number of devices, components and / or modules, etc. It should also be understood and recognized that various systems may include additional devices, components and / or modules, etc., and / or may not include all of the devices, components, modules, etc. discussed in relation to the drawings.

[0034] Terms such as "examples," "examples," "aspects," and "examples" as used herein may not be interpreted as implying that any aspect or design described is superior or more advantageous than other aspects or designs. Terms used below, such as "part," "component," "module," "system," and "interface," generally refer to computer-related entities and may refer, for example, to hardware, a combination of hardware and software, or software.

[0035] Additionally, the terms “comprising” and / or “comprising” should be understood to mean that the relevant feature and / or component is present, but not to exclude the presence or addition of one or more other features, components and / or groups thereof.

[0036] Additionally, terms including ordinal numbers, such as first, second, etc., may be used to describe various components, but said components are not limited by said terms. These terms are used solely for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be named the second component, and similarly, the second component may be named the first component. The term "and / or" includes a combination of a plurality of related described items or any of a plurality of related described items.

[0037] Furthermore, in the embodiments of the present invention, all terms used herein, including technical or scientific terms, unless otherwise defined, have the same meaning as generally understood by those skilled in the art to which the present invention pertains. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in the embodiments of the present invention.

[0039] FIG. 1 illustrates a server system that performs a method for recommending dog clothing based on review data analysis according to an embodiment of the present invention.

[0040] A method for recommending dog clothing based on review data analysis, performed in a server system comprising one or more processors and one or more memories according to an embodiment of the present invention, wherein the server system includes a database in which actual total body lengths are stored according to a combination of clothing information, including a body circumference, body length, and neck circumference for each of a plurality of clothing products received from a seller, and a plurality of dog breed information, weight information, and body shape information, are stored; and the method for recommending dog clothing comprises: a review data collection step of collecting a plurality of review data including one or more of dog breed information, weight information, and image information of a dog from a buyer who has purchased the clothing products; and an actual length derivation step of deriving the total body length, body circumference, body length, and neck circumference identified in an image from the image information included in each of the plurality of review data, and deriving the actual body circumference, actual body length, and actual neck circumference of each buyer's dog based on the total body lengths and a plurality of actual total body lengths stored in the database. The method may include: a clothing information update step for deriving updated clothing information based on the actual torso circumference, actual torso length, and actual neck circumference for each of the plurality of review data, based on clothing information entered by the seller for the clothing product; and a clothing recommendation step for recommending the clothing product to the new buyer by comparing the actual torso circumference, actual torso length, and actual neck circumference for the new buyer's pet dog derived based on input information entered by the new buyer with the updated clothing information.

[0042] The server system (1) according to one embodiment of the present invention may correspond to a computing readable recording medium comprising one or more processors and one or more memories. The server system (1) may correspond to a computing device that transmits and receives data and performs communication with terminals of external buyers and new buyers through a network.

[0044] As illustrated in FIG. 1, the server system (1) may be configured to include a review data collection unit (10), an actual length derivation unit (11), a clothing information update unit (12), a clothing recommendation unit (13), and a database (14).

[0046] The above database (14) can perform the function of storing and managing data necessary for performing the dog clothing recommendation method of the server system (1) and data generated by the dog clothing recommendation method.

[0047] Specifically, the database (14) can store initial clothing information including body circumference, body length, and neck circumference for each clothing product entered by the seller when first registering. Additionally, the database (14) can store standardized actual total body lengths that are statistically predicted to be possessed by the dogs of a group in the form of a mapping table corresponding to various combinations of body shape information including breed information, weight information, and one or more of underweight, normal, and overweight.

[0048] Furthermore, the above database (14) can store updated clothing information that includes size range information corrected by reflecting actual buyers' review data, which is subsequently generated by the clothing information update unit (12), so that it can be used when recommending clothing.

[0050] The above review data collection unit (10) can perform the function of obtaining data regarding a buyer's pet dog from multiple buyers who have purchased a clothing product. Specifically, the above review data collection unit (10) performs a review data collection step (S10) of collecting review data including dog breed information, weight information, and image information entered by the buyer. At this time, the image information collected may correspond to an image of the pet dog wearing the clothing product.

[0052] The above actual length derivation unit (11) can calculate the actual body dimensions of the dog in the image by analyzing the collected review data.

[0053] Specifically, the actual length derivation unit (11) can perform an actual length derivation step (S11) by using a pre-trained deep learning-based length derivation model to derive feature points and pixel lengths on an image, and matching them with the actual total body length stored in the database (14) to convert them into actual physical values ​​(cm) through ratio calculation. Through this, the actual body circumference, actual body length, and actual neck circumference of the reviewer's dog can be obtained without separate measurement.

[0054] A detailed explanation of the above length derivation model will be provided later in Fig. 3.

[0056] The above clothing information update unit (12) can perform the function of correcting the seller's initial clothing information by statistically analyzing a large amount of data regarding actual body circumference, actual body length, and actual neck circumference. Specifically, the above clothing information update unit (12) can perform a clothing information update step (S12) that calculates the average and standard deviation of the actual body circumference, actual body length, and actual neck circumference of dogs that have purchased the clothing product, and based on this, sets a size range that is actually suitable for the clothing and generates updated clothing information. The above updated clothing information can serve as an actual size standard that reflects the manufacturer's labeling error or the elasticity of the material.

[0058] The above clothing recommendation unit (13) can perform the function of selecting and providing optimal clothing to a new buyer. Specifically, the above clothing recommendation unit (13) can perform a clothing recommendation step (S13) by analyzing input information entered by a new buyer to derive the actual body circumference, actual body length, and actual neck circumference of the new dog and comparing them with previously generated updated clothing information, and determining whether the dimensions of the new dog fall within the appropriate range of the updated clothing information to recommend clothing.

[0060] According to one embodiment of the present invention, by updating clothing information based on actual body measurements obtained by reverse-calculating review data from multiple users who actually purchased the product, rather than relying on unilateral clothing size information provided by the seller, it is possible to achieve the effect of establishing realistic fit standards that reflect the manufacturer's measurement errors or material characteristics.

[0062] FIG. 2 illustrates the steps of performing a method for recommending dog clothing based on review data analysis according to an embodiment of the present invention.

[0063] The actual length derivation step according to one embodiment of the present invention comprises: a length calculation step in which image information included in each of the plurality of review data is input into a deep learning model that has been trained to derive values ​​corresponding to torso circumference, torso length, neck circumference, and total body length based on a plurality of feature points, thereby deriving torso circumference, torso length, neck circumference, and total body length corresponding to each of the plurality of image information; and a ratio calculation step in which, for each of the torso circumference, torso length, neck circumference, and total body length corresponding to each of the plurality of image information, a first ratio information regarding the ratio of torso circumference to total body length, a second ratio information regarding the ratio of torso length to total body length, and a third ratio information regarding the ratio of neck circumference to total body length are calculated. and may include an actual length calculation step of deriving a matching total body length that matches each of the plurality of review data among the actual total body lengths included in the database, and for each of the plurality of review data, calculating the actual body circumference, actual body length, and actual neck circumference of each dog in the review data based on the matching total body length, the first ratio information, the second ratio information, and the third ratio information.

[0064] Additionally, the clothing information update step according to one embodiment of the present invention may include: a statistical calculation step for calculating the average and standard deviation for each of the actual torso circumference, actual torso length, and actual neck circumference of the plurality of review data derived in the actual length derivation step; a range calculation step for calculating a range interval equal to a preset multiple of the standard deviation centered on the average for each of the actual torso circumference, actual torso length, and actual neck circumference, thereby deriving a first range interval for the actual torso circumference, a second range interval for the actual torso length, and a third range interval for the actual neck circumference; and an update information generation step for generating updated clothing information that updates the clothing information entered by the seller based on the first range interval, the second range interval, and the third range interval.

[0066] As illustrated in FIG. 2, the present invention may largely consist of a review data collection step (S10), an actual length derivation step (S11), a clothing information update step (S12), and a clothing recommendation step (S13).

[0068] First, in the review data collection step (S10) above, the server system (1) can receive review data from the terminals of users who have purchased clothing products. At this time, the received review data may include breed information, weight information, and image information of the dog wearing the clothing product for the buyer's pet dog.

[0069] Furthermore, in the above review data collection step (S10), each of the collected multiple review data can be input into a multi-modal model to analyze the text and image information included in the review data and to derive body shape information including one of overweight, standard weight, and underweight for the dog.

[0070] The above multimodal model may correspond to a model pre-trained based on a vast dataset of images of dogs of various breeds and body types, and labeling data of actual body measurements and obesity levels corresponding to each image.

[0071] Through this, the multimodal model can probabilistically calculate whether the dog is relatively underweight, normal weight, or overweight within the same breed and weight group, and derive body shape information that includes one of underweight, normal weight, or overweight, corresponding to the range with the highest probability. This is intended to find a more precise matching total body length by considering not only simple weight but also body shape when matching reference data from the database in the subsequent actual length derivation stage.

[0073] The above actual length derivation step (S11) is a process of converting collected unstructured image data into quantitative numerical data, and may specifically include a length derivation step (S110), a ratio calculation step (S111), and an actual length calculation step (S112).

[0075] In the above length derivation step (S110), a pre-trained deep learning-based length derivation model identifies external feature points including the nose tip, tail, neck, chest, etc. of a dog in the input image information, and based on this, outputs the total body length, body circumference length, body length, and neck circumference length, which are pixel unit lengths on the image.

[0076] Specifically, the length derivation model derives multiple feature points from image information and, based on these multiple feature points, can quantify and derive the total body length, which is the distance from the tip of the dog's head to the starting point of its tail; the body circumference, which is the width of the thickest part of the chest; the body length, which is the length of the back line from the point where the neck meets the back to the tail; and the neck circumference, which is the circumference of the neck. The values ​​derived at this time may correspond to lengths in pixel units dependent on image resolution or relative size values, rather than actual physical units.

[0078] In the above ratio calculation step (S111), in order to eliminate errors due to distance from the camera or zoom magnification, a first ratio information which is the ratio of the torso circumference to the total body length, a second ratio information which is the ratio of the torso length to the total body length, and a third ratio information which is the ratio of the neck circumference to the total body length can each be calculated.

[0080] In the actual length calculation step (S112), a matching total body length corresponding to the actual total body length that matches the dog breed information and weight information entered by each of the multiple buyers and the body shape information derived in the review data collection step (S10) can be extracted from the actual total body lengths stored in the database (14). By multiplying the matching total body length by the first ratio information, second ratio information, and third ratio information calculated in the ratio calculation step (S111), the actual torso circumference, actual torso length, and actual neck circumference of the buyer's pet dog can be finally calculated for each of the multiple buyers.

[0082] The above clothing information update step (S12) is a process of resetting the size standard of the clothing product by aggregating the actual size data of individual buyers, and may include, in detail, a statistical calculation step (S120), a range calculation step (S121), and an update information generation step (S122).

[0084] In the above statistical calculation step (S120), the average value and standard deviation for each of the actual torso circumference, actual torso length, and actual neck circumference derived from multiple accumulated review data for the clothing product can be calculated.

[0085] At this point, to prevent statistical distortion caused by extreme data, an additional step of removing outliers using the median and absolute median deviation can be performed to obtain the mean and standard deviation.

[0086] Specifically, the median and the median absolute deviation are calculated for each of the total actual torso circumference, actual torso length, and actual neck circumference data, and review data containing actual torso circumference, actual torso length, or actual neck circumference that has a value outside the preset multiple range of the median absolute deviation from each median is determined as an outlier and removed from the existing multiple review data groups, after which the final mean and standard deviation can be calculated only for the multiple review data from which outliers have been removed.

[0088] In the above interval calculation step (S121), a final range interval can be derived based on the average value and standard deviation for a plurality of review data from which outliers calculated in the above statistical calculation step have been removed.

[0089] Specifically, by setting a range equal to a preset multiple of the standard deviation centered on the average value of multiple review data from which outliers have been removed, a first range segment valid for actual body circumference, a second range segment valid for actual body length, and a third range segment valid for actual neck circumference can be derived, which are statistically valid ranges that include the majority of buyers' dogs.

[0091] In the above update information generation step (S122), the torso circumference, torso length, and neck circumference included in the clothing information initially entered by the seller are updated with the previously derived first range section, second range section, and third range section, thereby generating updated clothing information that reflects the actual wearing range, and can be stored in the above database (14).

[0093] Finally, in the clothing recommendation step (S13), when a new buyer who has not purchased the clothing product inputs one or more of the breed information, weight information, and image information of their pet dog, the server system (1) calculates the actual dimensions of the pet dog, namely the actual body circumference, actual body length, and actual neck circumference, and then determines whether the actual body circumference, actual body length, and actual neck circumference fall within the range of the updated clothing information to recommend a suitable clothing product.

[0094] Specifically, the above clothing recommendation step (S13) may include a new information collection step, a new length derivation step, a new ratio calculation step, a new actual length calculation step, and a recommendation step.

[0095] First, in the clothing recommendation step (S13), the server system (1) can perform a new information collection step in which the input information entered by the new buyer is input into the multimodal model to derive body shape information for the new buyer's pet dog.

[0096] Subsequently, a new length derivation step can be performed in which the image information entered by the new buyer is input into the length derivation model to measure the total body length, body circumference, body length, and neck circumference on the image, and a new ratio calculation step can be performed in which the first ratio information, second ratio information, and third ratio information for the new buyer's dog are each calculated based on the measured total body length, body circumference, body length, and neck circumference.

[0097] Afterwards, a new actual length calculation step can be performed by querying the database (14) based on the dog breed information, weight information, and body shape information included in the above input information, deriving the actual total body length, i.e., the matching total body length, and multiplying it by the previously calculated first ratio information, second ratio information, and third ratio information to calculate the actual body circumference, actual body length, and actual neck circumference of the new buyer's pet dog.

[0098] Finally, a recommendation step can be performed to provide a suitable clothing product to the new buyer by comparing whether the derived actual body circumference, actual body length, and actual neck circumference of the new buyer's dog fall within the range interval of the updated clothing information.

[0100] According to one embodiment of the present invention, by setting a statistical interval using the mean and standard deviation of dimension information derived from review data and performing a preprocessing process to remove outliers, it is possible to prevent information distortion caused by extreme review data and achieve the effect of suggesting a size range suitable for the majority of users.

[0101] In addition, when analyzing image information included in review data, the deep learning model automatically extracts actual dimensions based on ratios by linking feature points within the image with the database without the need for separate measurement tools. This allows for the conversion of a vast amount of review data into quantitative size data and systematic accumulation in the database without requiring reviewers to input dimensions separately.

[0103] FIG. 3 illustrates a plurality of feature points and length definitions for deriving the total body length, torso circumference, torso length, and neck circumference according to one embodiment of the present invention.

[0104] According to one embodiment of the present invention, the total body length corresponds to the length from a first feature point referring to the tip of the dog's head to a second feature point referring to the starting point of the tail, the body circumference corresponds to the circumference at a third feature point corresponding to the thickest part of the dog's chest, the body length corresponds to the length measured along the back line from a fourth feature point referring to the starting point where the dog's neck and back meet to a fifth feature point referring to the starting point of the tail, and the neck circumference corresponds to the circumference measured based on a sixth feature point corresponding to the lowest part of the neck where the dog's neck and shoulder are connected.

[0105] In addition, according to one embodiment of the present invention, the length derivation model is a deep learning model trained based on training data in which at least one of the total body length, torso circumference length, torso length, and neck circumference length of a corresponding dog is labeled on a plurality of dog images according to the combination of dog breed information, weight information, and body shape information, and can identify a first feature point, a second feature point, a third feature point, a fourth feature point, a fifth feature point, and a sixth feature point that serve as reference points for deriving the total body length, torso circumference length, torso length, and neck circumference length from the image information, and can derive the total body length, torso circumference length, torso length, and neck circumference length based on the derived plurality of feature points.

[0107] The deep learning-based length derivation model used in the present invention can improve the accuracy of the analysis by setting anatomically clearly identifiable feature points rather than ambiguous regions when analyzing the body structure of a pet dog.

[0108] First, as a reference point for measuring the total body length, the tip of the nose, which is clearly distinguished from the background and located at the foremost point of the dog's body, can be set as the first feature point, and the point at the base of the tail, where the spine ends and the tailbone begins, can be set as the second feature point. As shown in FIG. 3 (a), the straight distance between the first feature point and the second feature point, or the curved distance along the spine line, can be defined as the total body length that serves as the reference scale of this system.

[0109] The body circumference (chest circumference) is one of the most important measurements when wearing clothing, and the point where the rib cage is most expanded, located just behind the dog's front legs, can be set as the third feature point. When analyzing images, it can be calculated based on the circumference of the cross-section passing through the corresponding third feature point.

[0110] Torso length is a factor determining the length of the garment and can be defined as the length measured along the spinal line from the fourth feature point, where the neck and shoulders meet, to the fifth feature point, where the tail begins. Here, the fifth feature point may be at the same location as the previously defined second feature point.

[0111] Finally, the neck circumference is the point where the neckline of the clothing item is located. The lowest point of the neck where the dog's neck and shoulder connect is set as the 6th feature point, and the circumference when wrapping the entire neck 360 degrees around this point can be measured. This corresponds to a position slightly below where the leash is worn, where the clothing naturally rests.

[0113] The length derivation model mentioned in FIGS. 1 and 2 may correspond to a model that has been pre-trained based on a dataset of tens of thousands of images of dogs of various breeds and body types. In each image of the training data used at this time, not only numerical information such as the actual total body length, body circumference, body length, and neck circumference of the dog included in the image, but also pixel coordinates from the first feature point to the sixth feature point defined in FIG. 3 may be stored together as metadata.

[0114] The above length derivation model can be designed to automatically detect multiple feature points within an input image that serve as criteria for measuring body length by learning the correlation between the external characteristics and skeletal structure of a pet dog based on the above training data. Specifically, the above length derivation model can be trained to predict the coordinates of multiple feature points, such as the tip of the nose, the starting point of the tail, and the connection point between the neck and shoulders, with high accuracy by analyzing visual patterns such as changes in pixel values, brightness, and texture within the image.

[0115] In summary, the above length derivation model can identify the first, second, third, fourth, fifth, and sixth feature points from the input image, respectively. Then, by calculating the distances between these identified feature points, the total body length, torso circumference, torso length, and neck circumference on the image can be finally derived.

[0117] According to one embodiment of the present invention, as review data accumulates, the standard information for clothing sizes is continuously updated and refined, thereby enabling the establishment of a virtuous cycle structure in which the accuracy of the recommendation system improves over time.

[0119] FIG. 4 illustrates an example of collecting a plurality of review data in a review data collection step according to an embodiment of the present invention.

[0121] As illustrated in FIG. 4(a), in the review data collection step (S10), the server system (1) can collect actual review data from multiple buyers who have experience purchasing and wearing clothing products. The review data collected at this time may include breed information and weight information for the pet dog entered by each of the multiple buyers, as well as image information of the pet dog wearing the clothing product. In the example of FIG. 4(a), review data from different buyers with information such as 'Poodle, 10kg', 'Bichon, 12kg', and 'Maltese, 8kg' is collected.

[0123] As illustrated in FIG. 4(b), the server system (1) can perform the step of deriving body shape information of each dog based on each of the collected multiple review data. Even if they are of the same breed and weight, the degree of obesity may differ from individual to individual, and this has a significant impact on actual body measurements (especially chest circumference). Therefore, the server system (1) can derive body shape information indicating whether the dog belongs to 'overweight', 'normal weight', or 'underweight' by utilizing a pre-trained multimodal model to comprehensively analyze the breed information, weight information, and image information included in each of the input multiple review data.

[0125] FIG. 5 illustrates an example of performing a length derivation step and a ratio calculation step from image information of review data according to an embodiment of the present invention.

[0127] FIG. 5(a) illustrates the length derivation step (S110), which is a detailed step of the actual length derivation step. The length derivation model can identify multiple external feature points of a dog on the image information input from each of the multiple review data, and measure the length in pixel units on the image information based on this. In the example of FIG. 5(a), the process of identifying the total body length, body circumference, etc., as pixel distances in the image information of 'Poodle, 10kg', which is one of the multiple review data, is shown. This process can be applied in the same way not only to one review data but also to the remaining collected review data.

[0129] FIG. 5(b) illustrates the ratio calculation step (S111), which is a detailed step of the actual length derivation step (S11). Since the previously derived pixel length is an absolute value that may vary depending on the shooting distance or zoom magnification, a process of converting it into unique ratio information can be performed. In the example of FIG. 5(b), it can be seen that the first ratio information, which is the ratio of the torso circumference to the total body length, is calculated as 1:0.3, the second ratio information, which is the ratio of the torso length to the total body length, is calculated as 1:0.6, and the third ratio information, which is the ratio of the neck circumference to the total body length, is calculated as 1:0.2. This process can be applied in the same way not only to a single review data but also to the remaining collected review data.

[0131] FIG. 6 illustrates an example of performing an actual length calculation step by referring to a database according to an embodiment of the present invention.

[0133] In the actual length calculation step (S112) illustrated in FIG. 6, the server system (1) can perform the process of querying the database (14) using the breed information and weight information of each of the plurality of review data and the body shape information derived through the multimodal model in the preceding review data collection step (S10) as key values. In the database (14), standardized actual total body length data according to various breeds, weight ranges, and body shape conditions is stored in the form of a mapping table.

[0134] For example, in the example of FIG. 6, it can be seen that the actual total body length corresponding to the 'Poodle' breed, the weight range of '8~10kg', and the 'normal' body type is defined as '15cm'. The server system (1) can determine '15cm', which corresponds to the review data (Poodle, 10kg, normal), as the matching total body length of the corresponding pet dog. The matching total body length serves as a standard scale for converting the ratio information on the image into an actual physical length.

[0136] Subsequently, the server system (1) calculates the final actual dimensions by multiplying the first ratio information, the second ratio information, and the third ratio information calculated in the ratio calculation step (S111) of FIG. 5, respectively, by the determined matching total body length (15 cm). As shown at the bottom of FIG. 6, the actual torso circumference length is calculated as 4.5 cm by multiplying the matching total body length of 15 cm by the first ratio information of 0.3, the actual torso length is calculated as 9 cm by multiplying the 15 cm by the second ratio information of 0.6, and the actual neck circumference length is calculated as 3 cm by multiplying the 15 cm by the third ratio information of 0.2.

[0137] This process can be performed individually for each of the collected review data.

[0139] FIG. 7 illustrates an example of performing a statistical calculation step and a segment calculation step based on the actual torso circumference, actual torso length, and actual neck circumference derived for each of the plurality of review data according to one embodiment of the present invention.

[0141] FIG. 7(a) illustrates the statistical calculation step (S120), which is a detailed process of the clothing information update step (S12). In the statistical calculation step (S120), the server system (1) can calculate the average and standard deviation for the actual torso circumference, actual torso length, and actual neck circumference data for the multiple buyers derived earlier.

[0142] Looking at the table in Fig. 7 (a), review data from buyers of various dog breeds and body types, such as 'Poodle (10kg)', 'Bichon (12kg)', 'Maltese (8kg)', and 'Jindo (14kg)', has been accumulated.

[0143] In this case, for 'Jindo (14kg)', the actual torso circumference (10), actual torso length (20), and actual neck circumference (9) show significantly larger values ​​compared to other data. To prevent average distortion caused by such outliers, the above statistical calculation step (S120) may further include a step of calculating the median value and the median absolute deviation representing the dispersion of the actual torso circumference, actual torso length, and actual neck circumference, and using the median value and median absolute deviation, determining data that deviates from a preset median absolute deviation multiple as statistical outliers and excluding them from the aggregation target.

[0144] In the example of Fig. 7(a), for instance, when the median absolute deviation multiple is 3, the median is calculated as 5.25 when the actual torso circumference data values ​​are 4, 4.5, 6, and 10, and the median absolute deviation, which is the median of the difference between each data point and the median, is calculated as 1. At this time, the difference between the torso circumference of 'Jindo' (10) and the median (5.25) is 4.75, which falls within the range of the median absolute deviation multiple of a median absolute deviation of 1, i.e. Since it exceeds 3, the review data for 'Jindo' is judged as an outlier and may be excluded from the existing multiple review data.

[0146] FIG. 7(b) illustrates an example of multiple review data ('Poodle', 'Bichon', 'Maltese') remaining after excluding an outlier ('Jindo') through the process of FIG. 7(a). After removing the outlier, the statistical calculation step (S120) can calculate the mean value and standard deviation for the multiple review data excluding the outlier.

[0147] In the example of Fig. 7(b), the average of the actual torso circumference recalculated after removing outliers is 5 and the standard deviation is 1, the average of the actual torso length is 10 and the standard deviation is 2, and the average of the actual neck circumference is 3.5 and the standard deviation is 0.5, respectively.

[0149] According to one embodiment of the present invention, by utilizing the median of dimension information derived from review data to preemptively filter out outliers and calculating the mean and standard deviation using only refined data, it is possible to prevent distortion of statistical information caused by a small number of extreme review data and secure a highly reliable reference indicator.

[0151] FIG. 8 illustrates an example of generating updated clothing information by performing a clothing information update step according to an embodiment of the present invention.

[0153] FIG. 8(a) illustrates a section calculation step (S121) that calculates a review correction section, which is the final size tolerance range, using the average value and standard deviation derived from the preceding statistical calculation step (S120).

[0154] In the example of Fig. 8(a), it is assumed that the standard deviation multiple is set to 1.5. First, for the actual torso circumference, based on a mean of 5 and a standard deviation of 1, the lower limit (Rmin) is , the upper limit (Rmax) is It is calculated as such, and the first range section is derived as the range section for the actual torso circumference length of '3.5~6.5'. In the same way, for the actual torso length, the second range section is derived as the range section for the actual torso length of '7~13', and for the actual neck circumference length, the third range section is derived as '2.75~4.25'.

[0156] FIG. 8(b) illustrates an update information generation step (S122) that generates updated clothing information by applying the first range section, the second range section, and the third range section.

[0157] In the above update information generation step (S122), the clothing information 'body circumference 5, body length 9, neck circumference 4' originally provided by the seller can be updated to updated clothing information 'body circumference 3.5~6.5, body length 7~13, neck circumference 2.75~4.25' calculated through statistical calculations reflecting actual buyers' data. The updated clothing information can be used as reference data to determine whether the clothing product is suitable for the new buyer's dog when a new buyer is introduced.

[0159] FIG. 9 illustrates an example of calculating the actual body circumference, actual body length, and actual neck circumference of a new buyer's dog based on input information of a new buyer according to an embodiment of the present invention.

[0161] FIG. 9 illustrates the flow of the clothing recommendation step (S13) being actually performed, and each part represents the new information collection step, new length derivation step, new ratio calculation step, and new actual length calculation step corresponding to the detailed processing steps of the clothing recommendation step (S13).

[0162] As illustrated at the top of Fig. 9, a new buyer who has not purchased clothing products can input information including one or more of the dog breed information, weight information, and image information through their terminal. In the new information collection step, the input information is received and input into a multimodal model to derive the body shape information of the corresponding dog.

[0164] As illustrated in the middle of FIG. 9, after deriving the body shape information, the new length derivation step and the new ratio calculation step, which are detailed steps of the clothing recommendation step (S13), can be performed.

[0165] Specifically, in the new length derivation step, similar to the length derivation step (S110), image information included in the input information is input into the length derivation model to derive the total body length, body circumference, body length, and neck circumference of the new buyer's dog.

[0166] And in the above new ratio calculation step, similar to the above ratio calculation step (S111), the ratios of the torso circumference, torso length, and neck circumference to the total body length can be derived as the first ratio information, the second ratio information, and the third ratio information. In the example of FIG. 9, the first ratio information is calculated as '1:0.3', the second ratio information as '1:0.6', and the third ratio information as '1:0.2'.

[0168] The bottom of FIG. 9 illustrates the new actual length calculation step. Similar to the actual length calculation step (S112), in the new actual length calculation step, the matching total body length that matches the input information and the body shape information among the actual total body lengths stored in the database (14) is derived, and the actual body circumference, actual body length, and actual neck circumference of the new buyer's dog can be derived by multiplying the previously calculated ratio information.

[0169] As shown at the bottom of Fig. 9, it can be confirmed that the actual total body length corresponding to the 'Beagle' breed, '10kg' weight, and 'normal' body type entered by the new buyer is '15cm', and this can be determined as the matching total body length.

[0170] In addition, the final actual dimensions can be calculated by multiplying the above-mentioned matching total body length by the previously calculated first ratio information, second ratio information, and third ratio information, respectively. In the example illustrated in FIG. 9, the actual torso circumference is calculated as 4.5 (15 times 0.3), the actual torso length as 9 (15 times 0.6), and the actual neck circumference as 3 (15 times 0.2).

[0172] FIG. 10 illustrates an example of performing a clothing recommendation step by comparing size information and updated clothing information for a new buyer's pet dog according to one embodiment of the present invention.

[0174] Figure 10(a) illustrates the final result calculated through the series of processes (new information collection stage, new length derivation stage, new ratio calculation stage, new actual length calculation stage) described in Figure 9. That is, it shows that as a result of performing deep learning analysis and database matching based on input information including 'Beagle', '10kg', and image information entered by a new buyer, the actual body circumference of the new dog was determined to be 4.5, the actual body length to be 9, and the actual neck circumference to be 3.

[0176] Figure 10(b) illustrates a new recommendation step included in the clothing recommendation step. In the new recommendation step, the actual body circumference, actual body length, and actual neck circumference of the calculated new dog are compared with the updated clothing information stored in the database (14) to determine whether the clothing product is suitable for the new buyer's dog.

[0177] The above updated clothing information can be derived based on review data from numerous buyers through the process illustrated in FIGS. 4 to 8. In the new recommendation step, the server system (1) can determine whether the actual body circumference, actual body length, and actual neck circumference of the new buyer's dog, '4.5, 9, 3', are included in the product-specific size ranges stored in the database (14).

[0178] In the example of Fig. 10 (b), for 'Product A', the body circumference range is set to '17~21', so the new dog's dimensions (4.5) are outside the range and can be determined to be unsuitable. On the other hand, for 'Product B', all dimensions (4.5, 9, 3) of the new dog are stably included within the body circumference range (3.5~6.5), body length range (7~13), and neck circumference range (2.75~4.25), so it can be determined to be a suitable product. Likewise, 'Product C' also satisfies all range conditions and is included in the recommendation list. The server system (1) can generate a recommendation list including 'Product B' and 'Product C' through this comparative analysis and provide it to the new buyer.

[0180] In summary, the clothing recommendation step calculates the actual body circumference, actual body length, and actual neck circumference of the new dog from the input information of the new buyer, and compares the actual body circumference, actual body length, and actual neck circumference with updated clothing information updated by multiple review data to provide the new buyer with clothing products suitable for the dog's body measurements.

[0182] FIG. 11 illustrates an example of extracting a first layer and a second layer from image information and calculating hair length according to an embodiment of the present invention.

[0183] According to one embodiment of the present invention, the input information includes one or more of the breed information, weight information, and image information of the new buyer's pet dog, and the clothing recommendation step may further include: a step of extracting a first layer corresponding to an external silhouette including the pet dog's fur and a second layer corresponding to an internal silhouette excluding the pet dog's fur from the image information included in the input information; and a step of calculating a fur length compressible by the pet dog's fur based on the distance difference between the first layer and the second layer.

[0185] For long-haired or double-coated breeds with long, thick fur, such as Pomeranians, Bichon Frises, and Poodles, there can be a significant discrepancy between their visual volume and actual body size.

[0186] When deriving length based on feature points of image information included in the input information in the new length derivation step included in the above clothing recommendation step, since this is basically a value measured based on the appearance in a fur-covered state, there is a high possibility that it may be over-measured compared to the actual body.

[0187] Since clothing is worn close to the skin with the fur compressed, calculating measurements without considering the volume of the fur can lead to the recommendation of clothing that is excessively large compared to the actual size required. Therefore, to calculate precise measurements, it may be necessary to determine and correct for the thickness of the fur separately.

[0189] To solve this, the length derivation model used in the present invention can integrate the function of extracting feature points as described in FIG. 3, as well as the function of precisely separating objects within an image at the pixel level.

[0190] To perform this function, the length derivation model may be pre-trained based on vast training data consisting of pairs of original image data of companion dogs of various breeds and hair lengths, and information on the entire external shape area including the hair and the actual skin area inside the hair, labeled for each of the images by experts or precision measuring equipment.

[0191] Specifically, the above length derivation model can comprehensively analyze subtle texture differences and light and shadow of hair that are difficult to distinguish with the human eye, along with learned body structure data, to extract a first layer corresponding to the external silhouette, which is the boundary line of the visible, thick hair, and a second layer corresponding to the internal silhouette, which is the boundary line of the actual skin hidden within the hair, as independent regions.

[0192] At this time, the second layer may correspond to a virtual skin line where the actual body is predicted to be located by back-calculating the sagging or density of the hair due to gravity by the length derivation model, or it may correspond to a skeletal structure generated based on the first to sixth feature points and other major joint feature points derived in Fig. 6.

[0194] As illustrated in FIG. 11, the length derivation model can calculate the difference in distance between the boundary line of the extracted first layer and the boundary line of the second layer. This difference in distance represents the thickness of the hair covering the area and can be defined as the compressible hair length that can be compressed and reduced when wearing clothing. The hair length may be calculated as an average value over the entire torso, or it may be calculated locally for the neck or chest area, which is sensitive to the fit of the clothing.

[0196] The present invention may perform a process of comparing the hair length calculated through the process described above in the description of FIG. 11 with a preset hair length threshold to determine whether it is a meaningful subject for correction. In this case, since the torso length measured along the back line, even if the hair is abundant, is a body part where volume distortion—such as length increase or decrease due to the hair—rarely occurs, it can be excluded from the hair length subtraction target and the existing value can be maintained.

[0198] FIG. 12 illustrates a process of calculating the body circumference excluding hair and the neck circumference excluding hair by comparing the hair length and the hair length threshold according to one embodiment of the present invention.

[0199] The clothing recommendation step according to one embodiment of the present invention may include: a step of calculating the actual torso circumference, actual torso length, and actual neck circumference based on the input information; a step of calculating the torso circumference excluding hair and the neck circumference excluding hair by subtracting the hair length from the actual torso circumference and the actual neck circumference, respectively, when the hair length exceeds a preset hair length threshold; and a step of providing the corresponding clothing product when the torso circumference excluding hair, the neck circumference excluding hair, and the actual torso length correspond to the range included in the updated clothing information, by comparing the torso circumference excluding hair, the neck circumference excluding hair, and the actual torso length.

[0201] As illustrated in FIG. 12, if the hair length does not exceed the hair length threshold (NO), the server system (1) may consider the dog as a breed that does not require hair correction (short-haired breed, etc.), and maintain the actual body circumference and actual neck circumference calculated in the new actual length calculation step as they are and transmit them to the recommendation step.

[0203] On the other hand, if the hair length exceeds the hair length threshold (YES), the server system (1) can determine that the dog is a long-haired breed with significant volume distortion due to hair and perform a correction operation. To reflect the physical phenomenon where clothing presses down on the hair and adheres to the skin, the value obtained by subtracting the hair length from the actual body circumference can be calculated as the body circumference excluding hair, and the value obtained by subtracting the hair length from the actual neck circumference can be calculated as the neck circumference excluding hair and newly defined.

[0204] Specifically, the torso circumference excluding hair can be calculated by reflecting the volume reduction corresponding to the hair length in the actual torso circumference, and the neck circumference excluding hair can be calculated by reflecting the volume reduction corresponding to the hair length in the actual neck circumference.

[0205] Finally, the above server system (1) can determine the calculated body circumference excluding hair and neck circumference excluding hair as the effective body measurements of the corresponding dog and transmit them as input values ​​for the recommendation stage.

[0207] To summarize, if the hair length is less than or equal to the hair length threshold (NO), the actual torso circumference, actual torso length, and actual neck circumference calculated in the new actual length calculation step are transmitted to the recommendation step; and if the hair length exceeds the hair length threshold (YES), the actual torso length calculated in the new actual length calculation step is maintained, and the hair-excluded torso circumference and hair-excluded neck circumference, obtained by subtracting the hair length from the actual torso circumference and actual neck circumference, can be transmitted to the recommendation step.

[0209] According to one embodiment of the present invention, when analyzing an image, the outer silhouette including the dog's fur and the inner silhouette excluding the fur are distinguished through layer analysis, and the length inflated by the fur is calculated and subtracted from the actual body measurements, thereby resolving the problem of clothing being loose or ill-fitting even for dog breeds with thick fur, and thus providing the effect of recommending an optimized size for the dog.

[0211] FIG. 13 illustrates an example of deriving the distance between the eyes and the ratio of the two eyes when the input information according to one embodiment of the present invention corresponds to a mixed dog.

[0212] According to one embodiment of the present invention, the input information includes one or more of the breed information, weight information, and image information of the new buyer's pet dog, and the clothing recommendation step may include: a step of deriving the distance between the eyes corresponding to a part with minimal change in skeletal size according to the body shape information and the pet dog's breed from the image information included in the input information when the breed information included in the input information corresponds to a mixed dog with two or more breeds; a step of deriving the ratio of the distance between the eyes to the ratio of the total body length; and a body length correction step of calculating a correction value based on the ratio of the average value of a plurality of actual total body lengths derived based only on the weight information and body shape information included in the input information in the database, and calculating the actual torso circumference, actual torso length, and actual neck circumference of the mixed dog, respectively, based on the correction value.

[0214] With the increase in households with pet dogs, the proportion of mixed-breed dogs—which possess characteristics of two or more breeds rather than being purebreds—is on the rise. In the case of purebreds, standard body shape data by breed is relatively well-established, allowing for high accuracy to be expected through database matching alone. However, for mixed-breed dogs, body proportions can differ significantly from the standard body shape depending on the parent breeds or factors such as atavism. For example, even among mixed-breed dogs weighing the same 10kg, some may have long legs and a short body, while others may have a very long body like a Dachshund. Therefore, for mixed-breed dogs, it is difficult to derive accurate standard lengths from databases based solely on weight and body shape information.

[0216] To solve this, the present invention can utilize the distance between the eyes as a new correction reference point for mixed-breed dogs, as this part of the body has the least variation in skeletal structure and hardly changes in length even when gaining or losing weight.

[0217] As illustrated in FIG. 13, in the new ratio calculation step included in the clothing recommendation step, if the dog breed information input in the new information collection step corresponds to a mixed dog, the left eye and the right eye on the image can be identified as additional feature points through the length derivation model and the pixel distance between them can be measured.

[0218] Next, in the new ratio calculation step, the two-eye standard ratio, which is the ratio between the total body length derived in the new length derivation step and the length between the two eyes, can be calculated. In the example of FIG. 13, the two-eye standard ratio is derived as 1:0.2.

[0219] Since the above-mentioned two-eyed standard ratio is a value directly derived from the individual's unique traits without relying on a breed database, it can function as a key standard indicator for estimating the actual dimensions of mixed-breed dogs for which data is scarce.

[0221] The above-mentioned two-eyed standard ratio can be used as a standard measure to determine the specificity of the body proportions of the mixed dog and to accurately calculate the actual body length in the subsequent body length correction step.

[0222] The above body length correction step is a process for correcting the atypical body proportions of a mixed-breed dog, and corresponds to a step of determining the final actual body circumference, actual body length, and actual neck circumference by cross-referencing the average of the total body length in the database with the individual's unique two-eyed standard ratio derived earlier. In other words, it corresponds to a process that improves precision in deriving the body dimensions of a pet dog by correcting the length by reflecting the skeletal proportion characteristics of the individual rather than simply relying on average values, and a detailed explanation will be provided later in FIGS. 14 and 15.

[0224] FIG. 14 illustrates an example of calculating the total body length based on both eyes and the average total body length, respectively, to perform a body length correction step according to an embodiment of the present invention.

[0225] According to one embodiment of the present invention, the database stores the actual distance between the eyes, which is the actual length of the distance between the eyes corresponding to the combination of the weight information and the body shape information, and the body length correction step may include: a step of deriving the actual distance between the eyes that matches the weight information and the body shape information by referring to the database, and deriving the total body length based on the eyes by applying the ratio of the two eyes to the actual distance between the eyes; a step of calculating the average total body length, which is the average of the total body lengths of all dog breeds corresponding to the weight information and the body shape information by referring to the database; and a step of, when the difference between the average total body length and the total body length based on the eyes exceeds a preset correction threshold, calculating the corrected total body length by applying a weight based on the difference value to the total body length based on the eyes, and calculating the actual torso circumference, actual torso length, and actual neck circumference based on the corrected total body length.

[0227] In order to obtain accurate body measurements of a mixed dog for which standard data is lacking, the body length correction step may perform a process of comparing the two-eye standard total body length based on the two-eye standard ratio with the average total body length, which is the average value of the actual total body length stored in the database (14).

[0228] This comparison process serves as a criterion for determining whether the mixed dog falls within the range of normal body proportions or has a unique body shape with an unusually long or short waist, and by adjusting the weights accordingly, it is possible to make precise body measurement corrections even for mixed dogs for which standard data is unavailable.

[0230] As illustrated in FIG. 14 (a), in the body length correction step, an actual distance between the eyes value of 5 cm corresponding to “10 kg, normal body type” can be derived from the database (14). At this time, the derived actual distance between the eyes is skeletal information that hardly changes even if weight is gained or lost, so it can serve as a reliable reference point with little variation according to weight class.

[0231] The above body length correction step takes the actual distance between the two eyes, 5 cm, as the reference value and applies the two-eye standard ratio of 1:0.2 derived in the preceding novel ratio derivation step illustrated in FIG. 13 to the two-eye standard total body length It can produce.

[0233] Additionally, as illustrated in Fig. 14 (b), the body length correction step can calculate an average total body length of 16 cm, which is the average of the actual total body length data of all dog breeds corresponding to “10 kg, normal body type” in the database (14). This is length information that is universally possessed by dogs of the weight class to which the mixed dog belongs, and may correspond to a statistical standard value in which individual specificity is excluded.

[0235] FIG. 15 illustrates the process of calculating the final actual torso circumference, actual torso length, and actual neck circumference based on the difference value and correction threshold value according to one embodiment of the present invention.

[0237] In the above body length correction step, the difference between the previously calculated total body length based on both eyes and the average total body length is calculated, and if the difference is smaller than a preset correction threshold, it is determined that the mixed dog falls within the general body type category, and the above average total body length can be determined as the corrected total body length.

[0238] On the other hand, if the difference value exceeds a preset correction threshold, the individual is determined to have a unique body structure, and the final corrected total body length can be derived by assigning a higher weight to the value derived from the individual's inherent proportion.

[0240] As illustrated in FIG. 15, in the body length correction step, the total body length based on both eyes (25 cm) and the average total body length (16 cm) are calculated, and the difference value can be compared with a preset correction threshold value.

[0241] For example, if the above correction threshold is 3cm, the difference value of 9cm exceeds the correction threshold, so in the above body length correction step, this dog can be identified as having a unique body structure with an unusually long waist compared to a typical 8-10kg dog (e.g., Welsh Corgi or Dachshund mix, etc.).

[0242] Accordingly, in the above body length correction step, the corrected total body length can be calculated by assigning a higher weight than the average total body length of 16 cm to the total body length based on both eyes of 25 cm, which reflects the unique characteristics of the dog. Subsequently, the above body length correction step can calculate the actual torso circumference, actual torso length, and actual neck circumference by multiplying ratio information based on this corrected total body length.

[0244] For example, the corrected total body length by applying a weight of 0.8 to the above dual-eye-based total body length of 25cm and a weight of 0.2 to the average total body length of 16cm It can be calculated as.

[0245] And the actual torso circumference length ( ), actual torso length( ) and actual neck circumference( ) can be produced.

[0246] If the average total body length (16cm) was used without correction, the actual torso length is Although short clothes would have been recommended for the mixed dog with a long waist based on the calculation, by undergoing the correction process of the present invention, a measurement of 13.92 cm can be obtained that takes into account the long waist.

[0248] To summarize, in the clothing recommendation step, if the dog breed information included in the input information corresponds to a mixed dog, the server system (1) performs a process of deriving the distance between eyes and the ratio of the two eyes, which are the unique skeletal traits of the dog, through the length derivation model in the new ratio calculation step (Fig. 13), and in the body length correction step, performs a process of calculating the total body length based on the two eyes and the average total body length, which is the average in the database, respectively (Fig. 14). Subsequently, in the body length correction step, by analyzing the difference between the two values ​​and applying different weights depending on whether they exceed a preset correction threshold value to determine the final corrected total body length (Fig. 15), specific body dimensions can be derived even for mixed dogs that deviate from standard data.

[0249] The actual body circumference, actual body length, and actual neck circumference of the mixed breed dog, which are finally derived through the process described in FIGS. 13 to 15, are subsequently transmitted to the recommendation step, and the server system (1) can select clothing products suitable for mixed breed dogs by matching them with the seller's clothing products based on the actual body circumference, actual body length, and actual neck circumference.

[0251] According to one embodiment of the present invention, in the case of mixed-breed dogs for which it is difficult to directly apply standard body length information from a database, the distance between the eyes, which undergoes less change due to skeletal growth and is less affected by fur, is used as a reference point to correct the total body length, thereby enabling the effect of estimating dimensions with high accuracy even for various dog breeds for which standardized data is unavailable.

[0253] FIG. 16 schematically illustrates the internal configuration of a computing device according to one embodiment of the present invention.

[0254] The server system illustrated in FIG. 1 described above may include the components of the computing device (11000) illustrated in FIG. 16.

[0255] As illustrated in FIG. 16, the computing device (11000) may include at least one processor (11100), memory (11200), peripheral interface (11300), input / output subsystem (I / O subsystem) (11400), power circuit (11500), and communication circuit (11600). In this case, the computing device (11000) may correspond to the server system illustrated in FIG. 1.

[0256] The memory (11200) may include, for example, high-speed random access memory, a magnetic disk, SRAM, DRAM, ROM, flash memory, or non-volatile memory. The memory (11200) may include software modules, instruction sets, or various other data required for the operation of the computing device (11000).

[0257] At this time, access to memory (11200) from other components, such as the processor (11100) or peripheral device interface (11300), can be controlled by the processor (11100).

[0258] The peripheral device interface (11300) can connect input and / or output peripheral devices of the computing device (11000) to the processor (11100) and memory (11200). The processor (11100) can perform various functions for the computing device (11000) and process data by executing software modules or instruction sets stored in the memory (11200).

[0259] The input / output subsystem can connect various input / output peripherals to the peripheral interface (11300). For example, the input / output subsystem may include a controller for connecting peripherals such as a monitor, keyboard, mouse, printer, or, if necessary, a touchscreen or sensor to the peripheral interface (11300). According to another aspect, input / output peripherals may be connected to the peripheral interface (11300) without passing through the input / output subsystem.

[0260] The power circuit (11500) can supply power to all or part of the components of the terminal. For example, the power circuit (11500) may include one or more power sources such as a power management system, a battery or alternating current (AC), a charging system, a power failure detection circuit, a power converter or inverter, a power status indicator, or any other components for power generation, management, and distribution.

[0261] The communication circuit (11600) can enable communication with another computing device using at least one external port.

[0262] Alternatively, as described above, the communication circuit (11600) may enable communication with other computing devices by including an RF circuit and transmitting and receiving an RF signal, also known as an electromagnetic signal.

[0263] The embodiment of FIG. 16 is merely an example of a computing device (11000), and the computing device (11000) may have some components shown in FIG. 16 omitted, additional components not shown in FIG. 16 added, or a configuration or arrangement that combines two or more components. For example, a computing device for a communication terminal in a mobile environment may include a touchscreen or sensors, etc., in addition to the components shown in FIG. 16, and the communication circuit (11600) may include a circuit for RF communication of various communication methods (WiFi, 3G, LTE, Bluetooth, NFC, Zigbee, etc.). Components that can be included in the computing device (11000) may be implemented as hardware, software, or a combination of both hardware and software, including one or more integrated circuits specialized for signal processing or applications.

[0264] Methods according to embodiments of the present invention may be implemented in the form of program instructions that can be executed through various computing devices and recorded on a computer-readable medium. In particular, the program according to the present embodiment may be configured as a PC-based program or an application dedicated to a mobile terminal. An application to which the present invention is applied may be installed on a computing device (11000) through a file provided by a file distribution system. For example, the file distribution system may include a file transmission unit (not shown) that transmits the file in response to a request from the computing device (11000).

[0266] The device described above may be implemented as a hardware component, a software component, and / or a combination of a hardware component and a software component. For example, the device and components described in the embodiments may be implemented using one or more general-purpose or special-purpose computers, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA), a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing unit may execute an operating system (OS) and one or more software applications executed on said operating system. Additionally, the processing unit may access, store, manipulate, process, and generate data in response to the execution of the software. For ease of understanding, the processing unit may be described as being used as a single unit, but those skilled in the art will understand that the processing unit may include multiple processing elements and / or multiple types of processing elements. For example, the processing unit may include multiple processors or one processor and one controller. Additionally, other processing configurations, such as parallel processors, are also possible.

[0267] Software may include computer programs, code, instructions, or a combination of one or more of these, and may configure a processing unit to operate as desired or command the processing unit independently or collectively. Software and / or data may be permanently or temporarily embodied in any type of machine, component, physical device, virtual equipment, computer storage medium or device, or transmitted signal wave so as to be interpreted by the processing unit or to provide instructions or data to the processing unit. Software may be distributed over networked computing devices and stored or executed in a distributed manner. Software and data may be stored on one or more computer-readable recording media.

[0268] The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc., either alone or in combination. The program instructions recorded on the medium may be those specifically designed and configured for the embodiment, or they may be those known and available to those skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical recording media such as CD-ROMs and DVDs; magneto-optical media such as floptical disks; and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, and flash memory. Examples of program instructions include machine code, such as that generated by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc. The hardware devices described above may be configured to operate as one or more software modules to perform the operation of the embodiment, and vice versa.

[0270] Although the embodiments have been described above with reference to limited examples and drawings, those skilled in the art can make various modifications and variations from the description above. For example, appropriate results may be achieved even if the described techniques are performed in a different order than described, and / or if the components of the described system, structure, device, circuit, etc. are combined or assembled in a form different from described, or replaced or substituted by other components or equivalents. Therefore, other implementations, other embodiments, and equivalents to the claims are also included within the scope of the claims set forth below.

Claims

Claim 1 A method for recommending dog clothing based on review data analysis, performed on a server system comprising one or more processors and one or more memories, wherein the server system comprises a database storing actual total body lengths stored according to a combination of clothing information including body circumference, body length, and neck circumference for each of a plurality of clothing products received from a seller, and a plurality of dog breed information, weight information, and body shape information; and the method for recommending dog clothing comprises: a review data collection step of collecting a plurality of review data including one or more of dog breed information, weight information, and image information of a dog from a buyer who has purchased the clothing products; an actual length derivation step of deriving the total body length, body circumference, body length, and neck circumference identified in an image from the image information included in each of the plurality of review data, and deriving the actual body circumference, actual body length, and actual neck circumference of each buyer's dog based on the total body length and a plurality of actual total body lengths stored in the database; and the clothing information entered by the seller for the clothing products, the plurality of review data A clothing information update step for deriving updated clothing information based on the actual torso circumference, actual torso length, and actual neck circumference for each; and a clothing recommendation step for recommending clothing products to the new buyer by comparing the actual torso circumference, actual torso length, and actual neck circumference of the new buyer's dog derived based on input information entered by the new buyer with the updated clothing information; wherein the input information includes one or more of the breed information, weight information, and image information of the new buyer's dog, and the clothing recommendation step comprises, when the breed information included in the input information corresponds to a mixed dog with two or more breeds, a step of deriving the distance between the eyes corresponding to the length of a part that reflects the unique traits of the dog and has minimal changes in skeletal size according to the body shape information and the dog's breed from the image information included in the input information; and a step of deriving the ratio of the distance between the eyes to the ratio of the total body length.A method for recommending dog clothing, comprising: a body length correction step of calculating a correction value based on the two-eyed standard ratio for a plurality of average values ​​of actual total body lengths derived based only on the weight information and body shape information included in the input information in the database, and calculating the actual torso circumference, actual torso length, and actual neck circumference of the mixed dog, respectively, based on the correction value.; Claim 2 A method for recommending dog clothing according to claim 1, wherein the clothing information update step comprises: a statistical calculation step for calculating an average and a standard deviation for each of the actual body circumference, actual body length, and actual neck circumference of the plurality of review data derived in the actual length derivation step; a range calculation step for calculating a range interval equal to a preset multiple of the standard deviation centered on the average for each of the actual body circumference, actual body length, and actual neck circumference, to derive a first range interval for the actual body circumference, a second range interval for the actual body length, and a third range interval for the actual neck circumference; and an update information generation step for generating updated clothing information that updates the clothing information entered by the seller based on the first range interval, the second range interval, and the third range interval. Claim 3 In claim 1, the actual length derivation step comprises: a length calculation step of deriving the torso circumference, torso length, neck circumference, and total body length corresponding to each of the plurality of image information by inputting image information included in each of the plurality of review data into a deep learning model trained to derive values ​​corresponding to torso circumference, torso length, neck circumference, and total body length based on a plurality of feature points; and a ratio calculation step of calculating, for each of the torso circumference, torso length, neck circumference, and total body length corresponding to each of the plurality of image information, a first ratio information regarding the ratio of the torso circumference to the total body length, a second ratio information regarding the ratio of the torso length to the total body length, and a third ratio information regarding the ratio of the neck circumference to the total body length. A method for recommending dog clothing, comprising: a step of calculating actual lengths, wherein a matching total body length is derived from among the actual total body lengths included in the database that matches each of the plurality of review data, and for each of the plurality of review data, the actual torso circumference, actual torso length, and actual neck circumference of each dog in the review data are calculated based on the matching total body length, the first ratio information, the second ratio information, and the third ratio information. Claim 4 A method for recommending dog clothing according to claim 1, wherein the total body length corresponds to the length from a first feature point referring to the top of the dog's head to a second feature point referring to the starting point of the tail, the body circumference corresponds to the circumference at a third feature point corresponding to the thickest part of the dog's chest, the body length corresponds to the length measured along the back line from a fourth feature point referring to the starting point where the dog's neck and back meet to a fifth feature point referring to the starting point of the tail, and the neck circumference corresponds to the circumference measured based on a sixth feature point corresponding to the lowest part of the neck where the dog's neck and shoulder are connected. Claim 5 A method for recommending dog clothing according to claim 1, wherein the input information includes one or more of the breed information, weight information, and image information of the new buyer's dog, and the clothing recommendation step further comprises: a step of extracting a first layer corresponding to an outer silhouette including the dog's fur and a second layer corresponding to an inner silhouette excluding the dog's fur from the image information included in the input information; and a step of calculating a fur length compressible by the dog's fur based on the distance difference between the first layer and the second layer. Claim 6 In claim 5, the clothing recommendation step comprises: a step of calculating the actual body circumference, actual body length, and actual neck circumference based on the input information; a step of calculating the body circumference excluding hair and the neck circumference excluding hair by subtracting the hair length from the actual body circumference and the actual neck circumference, respectively, when the hair length exceeds a preset hair length threshold; and a step of comparing the body circumference excluding hair, the neck circumference excluding hair, and the actual body length with the updated clothing information, and providing the corresponding clothing product when the body circumference excluding hair, the neck circumference excluding hair, and the actual body length correspond to the range included in the updated clothing information. Claim 7 delete Claim 8 A review data analysis-based dog clothing recommendation system comprising one or more processors and one or more memories, comprising: a database storing actual total body lengths stored according to a combination of clothing information including body circumference, body length, and neck circumference for each of a plurality of clothing products received from a seller, and a plurality of dog breed information, weight information, and body shape information; a review data collection unit collecting a plurality of review data including one or more of dog breed information, weight information, and image information of a dog from a buyer who has purchased the clothing products; an actual length derivation unit deriving the total body length, body circumference, body length, and neck circumference identified in an image from the image information included in each of the plurality of review data, and deriving the actual body circumference, actual body length, and actual neck circumference of each buyer's dog based on the total body length and a plurality of actual total body lengths stored in the database; and the actual body circumference, actual body length, and actual neck circumference for the clothing information entered by the seller for the clothing products, and the plurality of review data A clothing information update unit that derives updated clothing information based on actual neck circumference; and a clothing recommendation unit that recommends clothing products to the new buyer by comparing the actual torso circumference, actual torso length, and actual neck circumference of the new buyer's dog derived based on input information entered by the new buyer with the updated clothing information; wherein the input information includes one or more of the breed information, weight information, and image information of the new buyer's dog, and the clothing recommendation unit, when the breed information included in the input information corresponds to a mixed dog with two or more breeds, derives the distance between the eyes corresponding to the length of a part that reflects the unique traits of the dog and has minimal changes in skeletal size according to the body shape information and the dog's breed from the image information included in the input information; and derives the ratio of the distance between the eyes to the ratio of the total body length.A pet dog clothing recommendation system that performs a body length correction step, wherein the average value of a plurality of actual total body lengths derived based only on the weight information and body shape information included in the input information in the database is corrected based on the two-eyed standard ratio, and the actual torso circumference, actual torso length, and actual neck circumference of the mixed dog are each calculated based on the corrected value.

Images