US 11,753,046 B2
System and method of predicting human interaction with vehicles
Samuel English Anthony, Cambridge, MA (US); Kshitij Misra, Cambridge, MA (US); and Avery Wagner Faller, Cambridge, MA (US)
Assigned to PERCEPTIVE AUTOMATA, INC., Boston, MA (US)
Filed by Perceptive Automata Inc., Boston, MA (US)
Filed on Sep. 7, 2021, as Appl. No. 17/468,516.
Application 17/468,516 is a continuation of application No. 16/828,823, filed on Mar. 24, 2020, granted, now 11,126,889.
Application 16/828,823 is a continuation of application No. 16/512,560, filed on Jul. 16, 2019, granted, now 10,614,344, issued on Apr. 7, 2020.
Application 16/512,560 is a continuation of application No. 15/830,549, filed on Dec. 4, 2017, granted, now 10,402,687, issued on Jan. 10, 2019.
Claims priority of provisional application 62/528,771, filed on Jul. 5, 2017.
Prior Publication US 2022/0138491 A1, May 5, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G06N 3/08 (2023.01); B60W 60/00 (2020.01); G06N 3/04 (2023.01); G08G 1/16 (2006.01); G08G 1/04 (2006.01); G05D 1/00 (2006.01); B60W 30/00 (2006.01); G06N 3/084 (2023.01); G06V 20/40 (2022.01); G06V 20/58 (2022.01); G06V 40/20 (2022.01); G06F 18/40 (2023.01); G06F 18/214 (2023.01); G06V 10/778 (2022.01); G06N 20/10 (2019.01); G06N 5/01 (2023.01)
CPC B60W 60/00274 (2020.02) [B60W 30/00 (2013.01); G05D 1/0088 (2013.01); G06F 18/214 (2023.01); G06F 18/41 (2023.01); G06N 3/04 (2013.01); G06N 3/08 (2013.01); G06N 3/084 (2013.01); G06V 10/7784 (2022.01); G06V 20/41 (2022.01); G06V 20/58 (2022.01); G06V 40/20 (2022.01); G08G 1/04 (2013.01); G08G 1/166 (2013.01); G05D 2201/0213 (2013.01); G06N 5/01 (2023.01); G06N 20/10 (2019.01)] 19 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
storing a plurality of images, each image displaying one or more users;
generating training data from the plurality of images, the generating comprising, for each image:
sending the image to a plurality of human observers, each human observer presented with a request to answer a question about a state of mind of a user in the image,
receiving, from each of the plurality of human observers, a response representing a judgment by the human observer of the state of mind of the user in the image,
generating summary statistics describing the state of mind of the user in the image based on the received responses from the plurality of human observers, and
storing the summary statistics in association with the image as part of the training data;
training a model using the training data, the model configured to receive an input image showing a user and predict summary statistics describing a state of mind of the user in the input image; and
executing the trained model to predict a state of mind of a user in a new image.