Method for determining authenticity of hanging images and deformation analysis of authenticity of item based on authentication score transgressing authenticity threshold
Systems and methods are directed to authenticating items using deformation analysis. An authentication system receives a request to authenticate an item. In response, the authentication system accesses two images of the item, whereby a first image shows the item hanging without a weight load and a second image shows the item hanging with the weight load. An image analyzer analyzes the two images to determine deformation differences of the item between the two images. The authentication system then determines an authentication score for the item by applying the deformation differences to a machine learning model trained with training data derived from authentic and counterfeit items. Based on the authentication score transgressing an authenticity threshold, an indication of authenticity of the item is provided.
1 . A method comprising:
receiving a request to authenticate an item;
accessing two images of the item, a first image showing the item hanging without a weight load and a second image showing the item hanging with the weight load;
analyzing, by an image analyzer, the two images to determine deformation differences of the item between the two images;
determining an authentication score for the item by applying the deformation differences to a machine learning model trained with training data derived from authentic and counterfeit items;
determining whether the authentication score transgresses an authenticity threshold; and
based on the authentication score transgressing the authenticity threshold, causing presentation of an indication of authenticity of the item.
2 . The method of claim 1 , wherein applying the deformation differences to the machine learning model comprises applying a vector of deformation differences between the two images to the machine learning model.
3 . The method of claim 1 , further comprising:
training the machine learning model using the training data derived from the authentic and counterfeit items, the training data including deformation differences at various locations of the authentic and counterfeit items.
4 . The method of claim 3 , wherein:
training the machine learning model includes training on a year or weight for each of the authentic items; and
the authentication probability is based in part on a year or weight of the item.
5 . The method of claim 3 , wherein the various locations comprise different locations on a handle of a handbag or a bottom of a handbag.
6 . The method of claim 1 , further comprising:
identifying a brand and model of the item by analyzing at least one of the images of the item or based on a serial number of the item.
7 . The method of claim 1 , wherein determining the authentication score further comprises:
identifying a time when the item was created or purchased; and
using the time as an additional feature to determine the authentication score.
8 . The method of claim 1 , wherein the two images are received from a client application running on a mobile device with camera capabilities for capturing the two images.
9 . The method of claim 1 , wherein the item is a handbag and the deformation differences comprise differences in a shape of a handle or a shape of a body of the handbag.
10 . The method of claim 1 , wherein causing presentation of the indication of authenticity comprises graphically displaying an authenticity badge on a listing for the item.
11 . A system comprising:
one or more processors; and
a memory storing instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
receiving a request to authenticate an item;
accessing two images of the item, a first image showing the item hanging without a weight load and a second image showing the item hanging with the weight load;
analyzing, by an image analyzer, the two images to determine deformation differences of the item between the two images;
determining an authentication score for the item by applying the deformation differences to a machine learning model trained with training data derived from authentic and counterfeit items;
determining whether the authentication score transgresses an authenticity threshold; and
based on the authentication score transgressing the authenticity threshold, causing presentation of an indication of authenticity of the item.
12 . The system of claim 11 , wherein applying the deformation differences to the machine learning model comprises applying a vector of deformation differences between the two images to the machine learning model.
13 . The system of claim 11 , wherein the operations further comprise:
training the machine learning model using the training data derived from the authentic and counterfeit items, the training data including deformation differences at various locations of the authentic and counterfeit items.
14 . The system of claim 13 , wherein:
training the machine learning model includes training on a year or weight for each of the authentic items; and
the authentication probability is based in part on a year or weight of the item.
15 . The system of claim 13 , wherein the various locations comprise different locations on a handle of a handbag or a bottom of a handbag.
16 . The system of claim 11 , wherein the operations further comprise:
identifying a brand and model of the item by analyzing at least one of the images of the item or based on a serial number of the item.
17 . The system of claim 11 , wherein determining the authentication score further comprises:
identifying a time when the item was created or purchased; and
using the time as an additional feature to determine the authentication score.
18 . The system of claim 11 , wherein the two images are received from a client application running on a mobile device with camera capabilities for capturing the two images.
19 . The system of claim 11 , wherein the item is a handbag and the deformation differences comprise differences in a shape of a handle or a shape of a body of the handbag.
20 . A machine-storage medium comprising instructions which, when executed by one or more processors of a machine, cause the machine to perform operations comprising:
receiving a request to authenticate an item;
accessing two images of the item, a first image showing the item hanging without a weight load and a second image showing the item hanging with the weight load;
analyzing, by an image analyzer, the two images to determine deformation differences of the item between the two images;
determining an authentication score for the item by applying the deformation differences to a machine learning model trained with training data derived from authentic and counterfeit items;
determining whether the authentication score transgresses an authenticity threshold; and
based on the authentication score transgressing the authenticity threshold, causing presentation of an indication of authenticity of the item.