Behavior recognition apparatus, learning apparatus, and method
First Claim
1. A behavior recognition apparatus comprising:
- a processor programmed to;
detect, based on a vehicle interior image obtained by photographing a vehicle interior, positions of a plurality of body parts of a person inside a vehicle in the vehicle interior image;
calculate distances between the detected positions of each of the plurality of body parts relative to each other, such that distances between all combinations of the plurality of body parts is calculated;
calculate a rank-order feature that ranks magnitudes the calculated distances between each combination of the detected positions of the plurality of body parts; and
discriminate a behavior of an occupant in the vehicle based on using the calculated rank order feature as an input to a discriminator learned in advance, which recognizes the behavior of the occupant inside the vehicle as an output of the discriminator.
1 Assignment
0 Petitions
Accused Products
Abstract
Provided is a behavior recognition apparatus, including a detection unit configured to detect, based on a vehicle interior image obtained by photographing a vehicle interior, positions of a plurality of body parts of a person inside a vehicle in the vehicle interior image; a feature extraction unit configured to extract a rank-order feature which is a feature based on a rank-order of a magnitude of a distance between parts obtained by the detection unit; and a discrimination unit configured to discriminate a behavior of an occupant in the vehicle using a discriminator learned in advance and the rank-order feature extracted by the feature extraction unit. Also provided is a learning apparatus to learn the discrimination unit.
9 Citations
16 Claims
-
1. A behavior recognition apparatus comprising:
-
a processor programmed to; detect, based on a vehicle interior image obtained by photographing a vehicle interior, positions of a plurality of body parts of a person inside a vehicle in the vehicle interior image; calculate distances between the detected positions of each of the plurality of body parts relative to each other, such that distances between all combinations of the plurality of body parts is calculated; calculate a rank-order feature that ranks magnitudes the calculated distances between each combination of the detected positions of the plurality of body parts; and discriminate a behavior of an occupant in the vehicle based on using the calculated rank order feature as an input to a discriminator learned in advance, which recognizes the behavior of the occupant inside the vehicle as an output of the discriminator. - View Dependent Claims (2, 3, 4)
-
-
5. A processor programmed to:
-
acquire positions of a plurality of body parts of a person inside a vehicle in a vehicle interior image obtained by photographing a vehicle interior, and a correct behavior taken by the person; calculate distances between the acquired positions of each of the plurality of body parts relative to each other, such that distances between all combinations of the plurality of body parts is calculated; calculate a rank-order feature that ranks magnitudes of the acquired distances between each combination of the detected positions of the plurality of body parts; and learn a discriminator that discriminates a behavior of an occupant in the vehicle based on using the rank-order feature and the correct behavior as inputs to the discriminator, which recognizes the behavior of the occupant inside the vehicle as an output of the discriminator. - View Dependent Claims (6, 7, 8)
-
-
9. A behavior recognition method comprising steps of:
-
detecting, based on a vehicle interior image obtained by photographing a vehicle interior, positions of a plurality of body parts of a person inside a vehicle in the vehicle interior image; calculating distances between the detected positions of each of the plurality of body parts relative to each other, such that distances between all combinations of the plurality of body parts is calculated; calculating a rank-order feature that ranks magnitudes of the calculated distances between each combination of the detected positions of the plurality of body parts; and discriminating a behavior of an occupant in the vehicle based on using the calculated rank order feature as an input to a discriminator learned in advance, which recognizes the behavior of the occupant inside the vehicle as an output of the discriminator. - View Dependent Claims (10, 11, 12)
-
-
13. A learning method comprising steps of:
-
acquiring positions of a plurality of body parts of a person inside a vehicle in a vehicle interior image obtained by photographing a vehicle interior and a correct behavior taken by the person inside the vehicle; calculating distances between the acquired positions of each of the plurality of body parts relative to each other, such that distances between all combinations of the plurality of body parts is calculated; calculating a rank-order feature that ranks magnitudes of the calculated distances between each combination of the detected positions of the plurality of body parts; and learning a discriminator that discriminates a behavior of an occupant in the vehicle based on using the rank-order feature and the correct behavior as inputs to the discriminator, which recognizes the behavior of the occupant inside the vehicle as an output of the discriminator. - View Dependent Claims (14, 15, 16)
-
Specification