Optimizing apparel combinations

Abstract

A apparel optimization system can present digital representations of apparel items in a user's wardrobe or at a store on an avatar built to resemble the user. The avatar can be sized to resemble the user, and the apparel items can be superimposed on the avatar. The system can provide outfit recommendations to the user based on the user's wardrobe, and visually display the recommended apparel items on the user's avatar. The system can also include providing future purchase recommendations based on seasonal, trend, and wardrobe composition data.

What is claimed is: 
     
         1 . A non-transitory, computer-readable medium containing instructions executed by at least one processor to perform stages for avatar-based apparel decisions, the stages comprising:
 storing, in a database, apparel data for a plurality of apparel items, the apparel data including size information of the apparel items and usage information for a plurality of users;   sizing an avatar for a first user among the plurality of users based on at least one of:
 a picture of a user in relation to a reference object; and 
 the size information for a first subset of the apparel items associated with the first user based on the usage information of each of the first subset of apparel items; and 
   scoring relationships between apparel items by a backend process based on the usage information;   determining a first apparel item to display on the avatar based on the scored relationships and the usage information for the first user;   displaying, on the user device, the first apparel item on the sized avatar.   
     
     
         2 . The non-transitory, computer-readable medium of  claim 1 , wherein determining the first apparel item to display includes:
 receiving selection of a second apparel item by the first user;   determining a match candidate for the second apparel item based on the first user's geographic location and usage information for the first user's apparel items having a compatible type to the second apparel item, wherein the match candidate is the first apparel item.   
     
     
         3 . The non-transitory, computer-readable medium of  claim 2 , wherein the first apparel item is chosen from apparel items that have not yet been owned by the first user, and wherein the stages further include
 providing a link to a purchase location for the first apparel item for selection by the first user.   
     
     
         4 . The non-transitory, computer-readable medium of  claim 2 , wherein the first apparel item is chosen from apparel items that are owned by the first user, and at least one of the owned apparel items is eliminated from selection based on recency of usage. 
     
     
         5 . The non-transitory, computer-readable medium of  claim 1 , wherein the first subset of apparel items are apparel items purchased by the first user. 
     
     
         6 . The non-transitory, computer-readable medium of  claim 1 , wherein determining the first apparel item to display includes weighting each of the first subset of apparel items based on a weather forecast at a geographic location of the first user, a calendar event associated with the first user, and a most recent date of usage for each of the apparel items in the subset. 
     
     
         7 . The non-transitory, computer-readable medium of  claim 1 , wherein using the size information for the first subset of apparel items to size the avatar includes:
 determining leg length data based on an average leg length of at least two apparel items worn by the first user within a recency threshold, the at least two apparel items being a pants type of apparel;   determining a waist size based on averaging waist size of the at least two apparel items; and   choosing a height and waist size of the avatar based on the determined leg length and waist size.   
     
     
         8 . The non-transitory, computer-readable medium of  claim 1 , the stages further comprising:
 receiving an image of the first user on a user device;   storing the image in association with the avatar; and   wherein displaying the avatar includes displaying a portion of the image on the avatar and superimposing onto the avatar at least the image a rendering of the first apparel item.   
     
     
         9 . The non-transitory, computer-readable medium of  claim 1 , the stages further comprising:
 receiving a selection from the first user for an outfit recommendation;   determining a geographic location of the first user;   receiving weather forecast information at the geographic location;   receiving calendar information for the user, the calendar information including an event type;   performing a first weighting for the scored plurality of apparel items based on the geographic location, weather forecast, and event type;   performing a second weighting for the plurality of apparel items based on past usage history of the scored plurality of apparel items by the first user;   determining a highest weighted primary apparel item;   determining a highest weighted complementary apparel item, based at least in part on a score of the complementary apparel item receiving additional weighting based on style and color compatibility with the highest weighted apparel item; and   display a first optimal apparel combination on the avatar, the first optimal apparel combination including the highest weighted primary apparel item and highest weighted complementary apparel item, wherein the first apparel item is the primary apparel item.   
     
     
         10 . The non-transitory, computer-readable medium of  claim 9 , wherein the score of the highest weighted complementary apparel item is determined by the backend process prior to the weighting and is relative to the highest weighted primary apparel item. 
     
     
         11 . The non-transitory, computer-readable medium of  claim 10 , further including displaying a second optimal apparel combination on a second avatar and contemporaneously with the first optimal apparel combination being displayed on the first avatar. 
     
     
         12 . The non-transitory, computer-readable medium of  claim 11 , wherein the first and second avatars are shared on a social network for soliciting input from the at least one other user. 
     
     
         13 . The non-transitory, computer-readable medium of  claim 1 , wherein the stages further include recommending a new apparel item in addition to the first apparel item based on at least one of:
 the user declining the first apparel item; and   age information of the first apparel item exceeding an age threshold;   wherein the new apparel item is selected based on a highest scored apparel item not yet owned by the user, wherein the scoring by the backend process is based at least in part on users having worn the highest scored apparel item together with apparel items that are owned by the first user.   
     
     
         14 . The non-transitory, computer-readable medium of  claim 13 , wherein the scoring by the backend process further includes:
 weighting apparel items differently for different geographic regions; and   weighting apparel items differently for different seasons of the year.   
     
     
         15 . A method for avatar-based apparel decisions, including:
 storing, in a database, apparel data for a plurality of apparel items, the apparel data including size information of the apparel items and usage information for a plurality of users;   sizing an avatar for a first user among the plurality of users based on at least one of:
 a picture of a user in relation to a reference object; and 
 the size information for a first subset of the apparel items associated with the first user based on the usage information of each of the first subset of apparel items; and 
   scoring relationships between apparel items by a backend process based on the usage information;   determining a first apparel item to display on the avatar based on the scored relationships and the usage information for the first user;   displaying, on the user device, the first apparel item on the sized avatar.   
     
     
         16 . The method of  claim 15 , wherein the stages further include recommending a new apparel item in addition to the first apparel item based on at least one of:
 the user declining the first apparel item; and   age information of the first apparel item exceeding an age threshold;   wherein the new apparel item is selected based on a highest scored apparel item not yet owned by the user, wherein the scoring by the backend process is based at least in part on users having worn the highest scored apparel item together with apparel items that are owned by the first user.   
     
     
         17 . The method of  claim 16 , wherein the scoring by the backend process further includes:
 weighting apparel items differently for different geographic regions; and   weighting apparel items differently for different seasons of the year.   
     
     
         18 . A system for avatar-based apparel decisions, including:
 a database that stores apparel data for a plurality of apparel items, the apparel data including size information of the apparel items and usage information for a plurality of users;   a server that executes a backend process that scores relationships between apparel items based on the usage information;   a front-end application installed on a computing device, the front-end application performing stages including:
 sizing an avatar for a first user among the plurality of users based on at least one of:
 a picture of a user in relation to a reference object; and 
 the size information for a first subset of the apparel items associated with the first user based on the usage information of each of the first subset of apparel items; and 
 
 determining a first apparel item to display on the avatar based on the scored relationships and the usage information for the first user; 
 displaying, on the computing device, the first apparel item on the sized avatar; 
 reporting, to the server, the first user's acceptance or denial of the first apparel item. 
   
     
     
         19 . The system of  claim 18 , wherein the front-end application performs further stages including:
 sending at least one apparel combination including the first apparel item to a second user for feedback;   displaying the feedback on the computing device; and   wherein the feedback is sent to the server for use in the scoring.   
     
     
         20 . The system of  claim 18 , wherein the front-end application performs further stages including:
 receiving an apparel recommendation from the server, the apparel recommendation including a second apparel item not owned by the first user, the recommendation being based on the second apparel item having a scored relationship with the first apparel item that exceeds a threshold, the scored relationship being based on usage of the first and second apparel items together by other users.