Browse Prior Art Database

DETERMINING VEHICLE OCCUPANT'S AGE

IP.com Disclosure Number: IPCOM000249854D
Publication Date: 2017-Apr-18
Document File: 2 page(s) / 188K

Publishing Venue

The IP.com Prior Art Database

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 62% of the total text.

1

DETERMINING VEHICLE OCCUPANT’S AGE

DETAILED DESCRIPTION

Vehicles can be equipped to operate in both autonomous and

occupant piloted mode. Vehicles can be equipped with computing

devices, networks, sensors and controllers to pilot the vehicle and to

determine maps of the surrounding real world including features such

as roads. A computing device can predict a vehicle occupant’s age based

on processing a video image that includes an image of the vehicle

occupant’s face to estimate the vehicle occupant’s age. Automatic age

estimation can be used to customize safety countermeasures for

vulnerable occupant protection. In addition to piloting a vehicle,

automatic age estimation has potential applications in law

enforcement, vehicle safety, security control, and human-computer

interaction.

Ranking-CNN, as discussed below with respect to Fig. 1, can

accurately and reliably predict a vehicle occupant’s age based on a

video image of the vehicle occupant’s face by aggregating binary

outputs from a series of trained basic CNNs. Fig. 1 is a diagram of a

ranking-CNN-based age estimation system 300 that includes multiple

basic CNNs 304, 308, 312, each trained to input a video image 302 and

estimate a vehicle occupant’s age using a deep age ranking approach.

Ranking-CNN-based age estimation system 300 includes multiple

CNN’s 304, 308, 312, where each CNN 304, 308, 312 is constructed in a

similar fashion to CNN 200, with a sequence of convolutional layers

C1, C3, C5, sub-sampling layers S2...