Gaitprint Imitation for AR Tracking Calibration and Pose Accuracy

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing augmented reality (AR) and virtual reality (VR) systems face challenges in accurately calibrating camera intrinsic parameters under varying operating conditions, leading to inefficiencies in pose estimation and resource utilization.

Innovation Solution

A method is introduced to transfer a user's gait pattern from one user to another to simulate AR content, generating training data that enhance the calibration of visual tracking systems, particularly through gaitprint imitation, thereby improving the accuracy of visual tracking and reducing computational resources.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional calibration methods are used for camera intrinsic parameters, then the system can operate with standard calibration procedures, but the pose estimation accuracy deteriorates under varying operating conditions

Engineering Contradiction:
Improvepose estimation accuracyVSAvoidcalibration adaptability to varying conditions
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The system performs preliminary gait pattern analysis and creates user-specific calibration models before actual AR operation. By pre-processing gait data and establishing personalized tracking parameters in advance, the system prepares optimized calibration settings that will be applied during runtime, improving both accuracy and adaptability without adding computational burden during operation.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system creates virtual copies of user gait patterns and transfers them to simulate AR content trajectories. By generating synthetic training data that replicates individual user movement characteristics, the system can pre-train visual tracking algorithms with diverse gait variations, enabling accurate pose estimation across different operating conditions without requiring extensive real-world calibration for each scenario.

Inventive Principle:
Principle #26Copying

2Measurement precision

If extensive calibration data and processing are used to improve accuracy, then pose estimation improves, but computational resources including processor cycles, memory usage, and power consumption increase

Engineering Contradiction:
Improvevisual tracking accuracyVSAvoidpower consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The system extracts only the essential gait pattern features and key trajectory parameters from extensive calibration data, separating the critical information needed for accurate tracking from redundant data. By isolating and retaining only the most important calibration elements, the system maintains high tracking accuracy while significantly reducing the computational resources required for processing and storing calibration information.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system transforms complex calibration datasets into optimized parameter sets by changing the representation format and dimensionality of calibration data. Through parameter optimization and dimensionality reduction techniques, the system converts large amounts of raw calibration information into compact, efficient parameter representations that maintain tracking accuracy while minimizing memory usage and processing requirements during operation.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP4341782B1Augmented reality data simulation with gaitprint imitation
Publication Date: 2025.11.19 SNAP INC
  • EP4341782B1 patent drawingFigure 1
  • EP4341782B1 patent drawingFigure 2
  • EP4341782B1 patent drawingFigure 3

AI summary

A method for transferring a gait pattern of a first user to a second user to simulate augmented reality content in a virtual simulation environment is described. In one aspect, the method includes identifying a gait pattern of a first user operating a first visual tracking system in a first physical environment, identifying a trajectory from a second visual tracking system operated by a second user in a second physical environment, the trajectory based on poses of the second visual tracking system over time, modifying the trajectory from the second visual tracking system based on the gait pattern of the first user, applying the modified trajectory in a plurality of virtual environments, and generating simulated ground truth data based on the modified trajectory in the plurality of virtual environments.