US 11,729,444 B2
System and methods for sensor-based audience estimation during digital media display sessions on mobile vehicles
Albert Stevens, St. Augustine, FL (US); and Ibrahim Atiya, Tallahassee, FL (US)
Assigned to Albert Stevens, St. Augustine, FL (US)
Filed by Adverge LLC, Dover, DE (US)
Filed on May 13, 2022, as Appl. No. 17/744,182.
Application 17/744,182 is a continuation in part of application No. 17/648,451, filed on Jan. 20, 2022.
Application 17/648,451 is a continuation in part of application No. PCT/IB2020/060733, filed on Nov. 15, 2020.
Prior Publication US 2022/0279226 A1, Sep. 1, 2022
Int. Cl. G06V 20/58 (2022.01); H04N 21/234 (2011.01); H04N 21/442 (2011.01); H04N 21/466 (2011.01); H04N 21/25 (2011.01); H04N 21/258 (2011.01); H04N 21/414 (2011.01); G06N 3/08 (2023.01); G06N 3/045 (2023.01)
CPC H04N 21/252 (2013.01) [G06N 3/045 (2023.01); G06N 3/08 (2013.01); H04N 21/25891 (2013.01); H04N 21/41422 (2013.01); H04N 21/44213 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A system for providing audience estimation for digital media display sessions displayed on a mobile vehicle, comprising:
non-transitory computer-readable storage media;
a processing system;
an interface system, including a communications interface;
one or more machine learning models;
program instructions stored on the computer-readable storage media that, when executed by the processing system, direct the processing system to:
receive, via the communications interface, object sensor data derived from one or more object sensors during a digital media display session;
apply an entity classifier produced by the one or more machine learning models to the object sensor data to determine a type of each unique audience entity;
for each unique audience entity, determine by the one or more machine learning models, from the object sensor data, a presence duration, a proximity, and a direction of the entity during the digital media display session; and
for each unique audience entity, compute an audience entity quality rating by executing the one or more machine learning models on inputs comprising the type, the presence duration, the proximity, and the direction of the audience entity, the one or more machine learning models outputting the audience entity quality rating.