Three-dimensional (3D) object detection and multi-agent behavior recognition using 3D motion data

A computer program product for object detection and behavior recognition system using three-dimensional motion data, the computer program product comprising computer-readable instructions stored on a non-transitory computer-readable medium that are executable by a computer having a processor for causing the processor to perform operations of:

• receiving three-dimensional (3D) motion data of a scene from at least one sensor;
• identifying at least a first object in the 3D motion data;
• extracting an object track of the first object, the object track being indicative of object motion in the scene over time;
• recognizing a behavior of the first object based on the object track;
• identifying at least a second object in the 3D motion data;
• extracting an object track of the second object, the object track being indicative of object motion in the scene over time;
• normalizing the object tracks of each of the first and the second objects to generate first and second normalized object tracks;
• and
• comparing the first and second normalized object tracks to calculate behavior differences and generate behavior results indicative of relative interactions between the first and second objects.

A method for object detection and behavior recognition system using three-dimensional motion data, the method comprising an act of causing a processor to execution instructions stored on a memory, such that upon execution of the instructions, the processor performs operations of:

• receiving three-dimensional (3D) motion data of a scene from at least one sensor;
• identifying at least a first object in the 3D motion data;
• extracting an object track of the first object, the object track being indicative of object motion in the scene over time;
• recognizing a behavior of the first object based on the object track;
• identifying at least a second object in the 3D motion data;
• extracting an object track of the second object, the object track being indicative of object motion in the scene over time;
• normalizing the object tracks of each of the first and the second objects to generate first and second normalized object tracks;
• and
• comparing the first and second normalized object tracks to calculate behavior differences and generate behavior results indicative of relative interactions between the first and second objects.