Systems and Methods for Automated Trade-In With Limited Human Interaction

Patent metadata

Field	Value
Publication number	US-2024378647-A1
Application number	US-202418781508-A
Country	US
Kind code	A1
Filing date	Jul 23, 2024
Priority date	Oct 22, 2019
Publication date	Nov 14, 2024
Grant date	—

Abstract

Aspects described herein may facilitate an automated trade-in of a vehicle with limited human interaction. A server may receive a request to begin a value determination of a vehicle associated with the user. The server may receive first data comprising: vehicle-specific identifying information, and multimedia content showing a first aspect of the vehicle. The user may be directed to place the vehicle within a predetermined staging area. The server may receive, from one or more image sensors associated with the staging area, second data comprising multimedia content showing a second aspect of the vehicle. The server may create a feature vector comprising the first data and the second data. The feature vector may be inputted into a machine learning algorithm corresponding to the vehicle-specific identifying information of the vehicle. Based on the machine learning algorithm, the server may determine a value of the vehicle.

First claim

What is claimed is:

1 . One or more non-transitory computer-readable medium storing computer instructions that, when executed by one or more processors, cause the one or more processors to perform steps comprising:
determining, based on feedback from one or more image sensors, that a vehicle associated with a user is within a predetermined area; receiving, from the one or more image sensors, first data comprising multimedia content showing a first aspect of damage to the vehicle associated with the user; creating a feature vector comprising the first data; identifying a machine learning algorithm corresponding to the vehicle associated with the user; and inputting the feature vector into the identified machine learning algorithm to determine a damage estimate to the first aspect of damage.

2 . The non-transitory computer-readable medium of claim 1 , wherein the computer instructions are further configured to:
receiving a request to begin a determination of the first aspect of damage; receiving a second data comprising multimedia content showing a second aspect of the vehicle associated with the user; and determining an initial damage estimate for the vehicle associated with the user.

3 . The non-transitory computer-readable medium of claim 2 , wherein the feature vector further comprises the second data.

4 . The non-transitory computer-readable medium of claim 1 , wherein the computer instructions are further configured to:
prior to the inputting the feature vector into the machine learning algorithm, identifying the machine learning algorithm based on vehicle-specific identifying information for the vehicle associated with the user.

5 . The non-transitory computer-readable medium of claim 1 , wherein the computer instructions are further configured to:
prior to the inputting the feature vector into the machine learning algorithm, training the machine learning algorithm using reference data of one or more aspects of a plurality of reference vehicles that are not associated with the user.

6 . The non-transitory computer-readable medium of claim 5 , wherein the computer instructions are further configured to train the machine learning algorithm by:
receiving, for each of the plurality of reference vehicles that are not associated with the user, reference vehicle-specific identifying information and reference data of the first aspect of a given reference vehicle of the plurality of reference vehicles; receiving, for each of the plurality of reference vehicles, an actual value of the given reference vehicle; creating, for each of the plurality of reference vehicles, a reference feature vector comprising the reference vehicle-specific identifying information and the reference data; associating, for each of the plurality of reference vehicles, the reference feature vector to the actual value of the given reference vehicle; and training the machine learning algorithm using the associated reference feature vectors to predict the actual value of the vehicle associated with the user based on the vehicle and the first data.

7 . The non-transitory computer-readable medium of claim 6 , wherein the computer instructions are further configured to:
determining a predictability for each of the one or more aspects of the reference vehicle for estimating the actual value of the given reference vehicle; and assigning, based on the determined predictability, a first weight to the first data.

8 . The non-transitory computer-readable medium of claim 1 , wherein the computer instructions are further configured to calibrate the one or more image sensors to produce the multimedia content based on a degree of illumination or a time within a diurnal cycle.

9 . The non-transitory computer-readable medium of claim 1 , wherein the first aspect comprises one or more of:
an exterior region of the vehicle associated with the user, or an interior region of the vehicle associated with the user.

10 . A method comprising:
determining, based on feedback from one or more image sensors, that a vehicle associated with a user is within a predetermined area; determining identifying information of the vehicle associated with the user receiving, from the one or more image sensors, first data comprising multimedia content showing a first aspect of damage to the vehicle associated with the user; creating a feature vector comprising the first data; identifying a machine learning algorithm corresponding to the vehicle associated with the user; and inputting the feature vector into the identified machine learning algorithm to determine a damage estimate to the first aspect of damage.

11 . The method of claim 10 , further comprising:
receiving a request to begin a determination of the first aspect of damage; receiving a second data comprising multimedia content showing a second aspect of the vehicle associated with the user; and determining an initial damage estimate for the vehicle associated with the user.

12 . The method of claim 11 , wherein the feature vector further comprises the second data.

13 . The method of claim 10 , further comprising:
prior to the inputting the feature vector into the machine learning algorithm, identifying the machine learning algorithm based on the identifying information of the vehicle associated with the user.

14 . The method of claim 10 , further comprising:
prior to the inputting the feature vector into the machine learning algorithm, training the machine learning algorithm using reference data of one or more aspects of a plurality of reference vehicles that are not associated with the user.

15 . The method of claim 14 , further comprising training the machine learning algorithm by:
receiving, for each of the plurality of reference vehicles that are not associated with the user, reference identifying information and reference data of the first aspect of a given reference vehicle of the plurality of reference vehicles; receiving, for each of the plurality of reference vehicles, an actual value of the given reference vehicle; creating, for each of the plurality of reference vehicles, a reference feature vector comprising the reference identifying information and the reference data; associating, for each of the plurality of reference vehicles, the reference feature vector to the actual value of the given reference vehicle; and training the machine learning algorithm using the associated reference feature vectors to predict the actual value of the vehicle associated with the user based on the vehicle and the first data.

16 . The method of claim 15 , further comprising:
determining a predictability for each of the one or more aspects of the reference vehicle for estimating the actual value of the given reference vehicle; and assigning, based on the determined predictability, a first weight to the first data.

17 . The method of claim 10 , further comprising calibrating the one or more image sensors to produce the multimedia content based on a degree of illumination or a time within a diurnal cycle.

18 . The method of claim 10 , wherein the first aspect comprises one or more of:
an exterior region of the vehicle associated with the user, or an interior region of the vehicle associated with the user.

19 . An apparatus comprising:
one or more processors; one or more image sensors; and a memory storing instructions that, when executed by the one or more processors, cause the apparatus to:
determine, based on feedback from the one or more image sensors, that

Assignees

• Capital One Services Llc

Inventors

• Price Micah
• Ghamsari Avid
• Dagley Geoffrey
• Tang Qiaochu
• Hoover Jason

Classifications

• G06V10/40

  Extraction of image or video features · CPC title

• G06V10/462

  Salient features, e.g. scale invariant feature transforms [SIFT] · CPC title

• G06V10/764

  using classification, e.g. of video objects · CPC title

• G06F18/24

  Classification techniques · CPC title

• G06V20/00

  Scenes; Scene-specific elements (control of digital cameras H04N23/60) · CPC title