Method, medium, and system for shopper integrity estimation in frictionless retail stores

1. A non-transitory computer-readable medium including instructions that when executed by a processor cause the processor to implement a frictionless shopping method based on estimated shopper integrity, the method comprising:

receiving, from at least a first of a plurality of sensors associated with a retail store, first sensed data indicating a selection of a first product by a customer, wherein the first of the plurality of sensors associated with the retail store includes a camera, and wherein the first sensed data includes image data captured by the camera;

determining, based on an analysis of the image data using a trained machine learning model, a first property associated with the first product;

issuing a first query to the customer with respect to the first property associated with the first product, wherein issuing the first query includes causing a plurality of choices to be displayed on a graphical user interface of a device associated with the customer;

in response to issuing the first query, receiving first information from the customer;

comparing the first information received from the customer with the first property associated with the first product;

determining, based on the comparison between the first information received from the customer with the first property associated with the first product, an integrity indicator associated with the customer;

receiving, from at least a second of the plurality of sensors associated with the retail store, second sensed data indicating a selection of a second product by the customer, wherein the second of the plurality of sensors associated with the retail store includes a second camera, and wherein the second sensed data includes second image data captured by the second camera;

determining, based on an analysis of the second image data using the trained machine learning model and based on the integrity indicator, at least one action to take in order to determine a second property associated with the second product; and

initiating the at least one action, wherein initiating the at least one action includes issuing a second query to the customer based on the integrity indicator associated with the first information received from the customer being accurate, and wherein issuing the second query includes causing a second plurality of choices to be displayed on the graphical user interface of the device associated with the customer.

2. The non-transitory computer-readable medium of claim 1 , wherein the first property is indicative of a product type associated with the first product.

3. The non-transitory computer-readable medium of claim 1 , wherein the first property is indicative of a quantity associated with the first product.

4. The non-transitory computer-readable medium of claim 1 , wherein the second property is indicative of a product type associated with the second product.

5. The non-transitory computer-readable medium of claim 1 , wherein the second property is indicative of a quantity associated with the second product.

6. The non-transitory computer-readable medium of claim 1 , wherein the first property and the second property relate to a common product characteristic.

7. The non-transitory computer-readable medium of claim 1 , wherein the first property and the second property relate to different product characteristics.

8. The non-transitory computer-readable medium of claim 1 , wherein the first information received from the customer includes a selection of at least one of a plurality of options presented to the customer.

9. The non-transitory computer-readable medium of claim 1 , wherein the first information received from the customer indicates at least one of a price, a selected quantity, or a product type for the first product.

10. The non-transitory computer-readable medium of claim 1 , wherein the at least one action further includes:

receiving second information from the customer in response to the second query; and

determining the second property based on the second information received from the customer.

11. The non-transitory computer-readable medium of claim 1 , wherein based on the integrity indicator being associated with the first information received from the customer being inaccurate, the at least one action further includes requiring the customer to scan a barcode associated with the second product.

12. The non-transitory computer-readable medium of claim 1 , wherein based on the integrity indicator being associated with the first information received from the customer being inaccurate, the at least one action further includes requiring the customer to capture an image of the second product.

13. The non-transitory computer-readable medium of claim 1 , wherein based on the integrity indicator being associated with the first information received from the customer being inaccurate, the at least one action further includes causing a store associate associated with the retail store to confirm the second property associated with the second product.

14. The non-transitory computer-readable medium of claim 1 , wherein based on the integrity indicator being associated with the first information received from the customer being inaccurate, the at least one action further includes requiring the customer to engage in a non-frictionless checkout process.

15. The non-transitory computer-readable medium of claim 1 , wherein at least one of the plurality of sensors associated with the retail store includes a pressure sensor.

16. The non-transitory computer-readable medium of claim 1 , wherein the method further comprises:

analyzing the second sensed data to determine a plurality of alternative plausible values for the second property associated with the second product; and

further basing the selection of the at least one action on the determined plurality of alternative plausible values.

17. The non-transitory computer-readable medium of claim 1 , wherein the method further comprises:

analyzing the second sensed data to determine a plurality of likelihoods, each likelihood of the plurality of likelihoods is associated with a different alternative plausible value for the second property associated with the second product; and

further basing the selection of the at least one action on a distribution of the determined plurality of likelihoods.

18. A frictionless shopping system, comprising:

at least one processor configured with access to instructions that when executed by the at least one processor cause the at least one processor to implement a frictionless shopping method based on estimated shopper integrity, the method comprising:

receiving, from at least a first of a plurality of sensors associated with a retail store, first sensed data indicating a selection of a first product by a customer, wherein the first of the plurality of sensors associated with the retail store includes a camera, and wherein the first sensed data includes image data captured by the camera;

determining, based on an analysis of the image data using a trained machine learning model, a first property associated with the first product;

issuing a first query to the customer with respect to the first property associated with the first product, wherein issuing the first query includes causing a plurality of choices to be displayed on a graphical user interface of a device associated with the customer;

in response to issuing the first query, receiving first information from the customer;

comparing the first information received from the customer with the first property associated with the first product;

determining, based on the comparison between the first information received from the customer with the first property associated with the first product, an integrity indicator associated with the customer;

receiving, from at least a second of the plurality of sensors associated with the retail store, second sensed data indicating a selection of a second product by the customer, wherein the second of the plurality of sensors associated with the retail store includes a second camera, and wherein the second sensed data includes second image data captured by the second camera;

determining, based on an analysis of the second image data using the trained machine learning model and based on the integrity indicator, at least one action to take in order to determine a second property associated with the second product; and

initiating the at least one action, wherein initiating the at least one action includes issuing a second query to the customer based on the integrity indicator associated with the first information received from the customer being accurate, and wherein issuing the second query includes causing a second plurality of choices to be displayed on the graphical user interface of the device associated with the customer.

19. A frictionless shopping method based on estimated shopper integrity, comprising:

receiving, from at least a first of a plurality of sensors associated with a retail store, first sensed data indicating a selection of a first product by a customer, wherein the first of the plurality of sensors associated with the retail store includes a camera, and wherein the first sensed data includes image data captured by the camera;

determining, based on an analysis of the image data using a trained machine learning model, a first property associated with the first product;

issuing a first query to the customer with respect to the first property associated with the first product, wherein issuing the first query includes causing a plurality of choices to be displayed on a graphical user interface of a device associated with the customer;

in response to issuing the first query, receiving first information from the customer;

comparing the first information received from the customer with the first property associated with the first product;

determining, based on the comparison between the first information received from the customer with the first property associated with the first product, an integrity indicator associated with the customer;

receiving, from at least a second of the plurality of sensors associated with the retail store, second sensed data indicating a selection of a second product by the customer, wherein the second of the plurality of sensors associated with the retail store includes a second camera, and wherein the second sensed data includes second image data captured by the second camera;

determining, based on an analysis of the second image data using the trained machine learning model and based on the integrity indicator, at least one action to take in order to determine a second property associated with the second product; and

initiating the at least one action, wherein initiating the at least one action includes issuing a second query to the customer based on the integrity indicator associated with the first information received from the customer being accurate, and wherein issuing the second query includes causing a second plurality of choices to be displayed on the graphical user interface of the device associated with the customer.