Back to results
US 11,599,811 B1Granted
Generating event results using auto-generated sensor data
AssigneeAmazon Technologies, Inc.
InventorsPaul Aksenti Savastinuk, Roman Talyansky, Michael Dillon, Eli Osherovich, Gopi Prashanth Gopal
Life of this patent
Estimating the term…
Abstract
This disclosure describes systems and techniques for detecting events, determining a result of each respective event using a first hypothesis source, and calculating a likelihood that a second (and/or additional) hypothesis source would determine the same result of the respective event. The calculated likelihood may then be used to be determine whether to request that the second hypothesis source determine the result of the event, determine an amount of resources of the second hypothesis source to use to make this determination, and/or like.
Claims23
A method comprising:
- receiving first sensor data indicative of a first event in a facility;
- sending at least one of the first sensor data or information for obtaining the first sensor data to a first hypothesis source;
- receiving an indication of a first hypothesis of a result of the first event from the first hypothesis source, the first hypothesis indicating that a first user performed a first action with respect to a first item;
- receiving a first confidence value from the first hypothesis source, the first confidence value indicating an estimation of accuracy of the first hypothesis;
- determining that the first confidence level is less than a first threshold;
- receiving data associated with the first event based at least in part on the sensor data;
- inputting the data into a model;
- receiving, based at least in part on the inputting, a second confidence value, the second confidence value indicating a likelihood that a second hypothesis source would determine the first hypothesis based at least in part on the first sensor data;
- determining that the second confidence value is greater than a second threshold;
- storing, in cart data associated with the first user, an indication that the first user performed the first action with respect to the first item;
- receiving second sensor data indicative of a second event;
- determining, based at least in part on the second sensor data, that the second event comprises the first user exiting the facility;
- and
- charging a payment instrument associated with the first user for items indicated in the cart data associated with the first user.
11 more elements
A method comprising:
- determining, by a first hypothesis source, a first hypothesis for a result of a first event based at least in part on first sensor data associated with the first event, the first hypothesis indicating that a first user performed a first action with respect to a first item;
- determining, by the first hypothesis source, a first confidence value indicating an estimation of accuracy of the first hypothesis;
- determining that the first confidence value fails to satisfy one or more criteria;
- receiving data associated with the first event, at least a portion of the data based at least in part on the first sensor data;
- inputting the data into a model;
- determining, from the model, a second confidence value indicating a likelihood that a second hypothesis source that is different than the first hypothesis source would generate a hypothesis consistent with the first hypothesis;
- storing, in cart data associated with the first user, an indication that the first user performed the first action with respect to the first item;
- determining, based at least in part on second sensor data, that the first user has exited the facility;
- and
- charging a payment instrument associated with the first user for items indicated in the cart data associated with the first user.
7 more elements
One or more computer-readable media storing computer-executable instructions that, when executed, cause one or more processors to perform acts comprising:
- determining, by a first hypothesis source, a first hypothesis for a result of a first event based at least in part on first sensor data associated with the first event, the first hypothesis indicating that a first user performed a first action with respect to a first item;
- determining, by the first hypothesis source, a first confidence value indicating an estimation of accuracy of the first hypothesis;
- determining that the first confidence value fails to satisfy one or more criteria;
- receiving data associated with the first event, at least a portion of the data based at least in part on the first sensor data;
- inputting the data into a model;
- determining, from the model, a second confidence value indicating a likelihood that a second hypothesis source that is different than the first hypothesis source would generate a hypothesis consistent with the first hypothesis;
- storing, in a cart data associated with the first user, an indication that the first user performed the first action with respect to the first item;
- determining, based at least in part on second sensor data, that the first user has exited the facility;
- and
- charging a payment instrument associated with the first user for items indicated in the cart data associated with the first user.
7 more elements
A method comprising:
- determining, by a first hypothesis source, a first hypothesis for a result of a first event based at least in part on first sensor data associated with the first event, the first hypothesis indicating that a first user physically picked up a first item;
- determining, by the first hypothesis source, a first confidence value indicating an estimation of accuracy of the first hypothesis;
- determining a second confidence value indicating a likelihood that a second hypothesis source that is different than the first hypothesis source would generate a hypothesis consistent with the first hypothesis;
- storing an indication that the first user physically picked up the first item;
- determining, based at least in part on second sensor data, that the first user has exited a facility;
- and
- charging a payment instrument associated with the first user for the first item.
4 more elements
Dependencies are inferred from the claim text.