Depth Sensor Subject Tracking via 3D Trunk and Head Modeling

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional video cameras face challenges in robust and reliable joint tracking due to ambient light changes, segmentation problems, and occlusion, especially when tracking subjects without the need for special equipment or calibration.

Innovation Solution

The use of depth sensors to generate image depth data, which allows for the creation of a three-dimensional model of a subject's joints without requiring explicit knowledge of the sensor's parameters or orientation, enabling automatic adaptation and articulation modeling based on the subject's position and orientation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If conventional video cameras are used to track subject movements, then the system is simple and does not require special equipment, but the tracking reliability deteriorates due to ambient light changes, segmentation problems, and occlusion

Engineering Contradiction:
Improvesystem simplicityVSAvoidtracking reliability
Core Design Contradiction:
Ease of manufactureVSReliability

Solution Approach 1:

The patent replaces conventional video camera optical systems with depth sensing technology that measures distance directly, substituting optical imaging with a depth measurement mechanism that is immune to lighting conditions and provides reliable joint tracking without segmentation problems

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent changes the measurement parameter from two-dimensional image coordinates to three-dimensional depth information, allowing the system to track subject joints reliably by using depth data that remains consistent regardless of ambient light changes or occlusion

Inventive Principle:
Principle #35Parameter changes

2Reliability

If depth sensors are used to track subject joints, then tracking reliability improves, but the device complexity increases due to sensor calibration requirements

Engineering Contradiction:
Improvejoint tracking reliabilityVSAvoidsensor calibration complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent implements self-calibration mechanisms where the depth sensor automatically determines its own parameters and orientation by analyzing the subject's body structure and joint positions, eliminating the need for manual calibration procedures and reducing device complexity

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent performs automatic sensor parameter determination and model adaptation as a preliminary step before tracking, allowing the system to pre-compute depth sensor characteristics and adjust the subject model accordingly, thereby simplifying subsequent tracking operations

Inventive Principle:
Principle #10Preliminary action

3Adaptability or versatility

If the system adapts to arbitrary subject positions and orientations, then the adaptability improves, but the computational complexity increases

Engineering Contradiction:
Improvesubject position adaptabilityVSAvoidmodeling computation complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent employs dynamic model adaptation where the subject's three-dimensional model is automatically adjusted in real-time based on detected joint positions and body orientation, allowing the system to handle arbitrary positions and orientations through continuous model updating rather than pre-computed static models

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS9330470B2Method and system for modeling subjects from a depth map
Publication Date: 2016.05.03 TAHOE RES LTD
  • US9330470B2 patent drawing
  • US9330470B2 patent drawing
  • US9330470B2 patent drawing

AI summary

A method for modeling and tracking a subject using image depth data includes locating the subject's trunk in the image depth data and creating a three-dimensional (3D) model of the subject's trunk. Further, the method includes locating the subject's head in the image depth data and creating a 3D model of the subject's head. The 3D models of the subject's head and trunk can be exploited by removing pixels from the image depth data corresponding to the trunk and the head of the subject, and the remaining image depth data can then be used to locate and track an extremity of the subject.