AR Point-of-Interest Tracking Through Dual-Pipeline Feature Mapping

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video object tracking techniques are limited by the use of a small number of trackable points, which compromises accuracy and quality in tracking objects, particularly facial features, and require significant computational resources and time.

Innovation Solution

A method and system that augment points of interest in augmented-reality videos by extracting feature points, determining a coordinate system, and using reference points to calculate and recalibrate location parameters, allowing for accurate tracking with minimal computational resources.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the number of trackable points is increased to improve tracking accuracy of object boundaries, then measurement precision is improved, but computational processing resources and time are significantly consumed

Engineering Contradiction:
Improvetracking accuracyVSAvoidcomputational processing efficiency
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent segments the tracking process into two distinct pipelines: a first pipeline that processes video frames in real-time using a limited set of trackable points, and a second pipeline that performs computationally intensive processing to determine precise location parameters of points of interest. This segmentation allows the system to maintain real-time processing capability while achieving high tracking accuracy through the coordinated work of both pipelines.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary mechanism where the first pipeline extracts feature points and transfers them to the second pipeline, which then calculates precise location parameters and transfers them back to the first pipeline for augmentation. This intermediary exchange of data between pipelines enables the system to achieve high measurement precision without requiring the first pipeline to perform computationally expensive calculations directly.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If the number of trackable points is increased to accurately track outlines and boundaries of facial features, then measurement precision is improved, but device complexity increases

Engineering Contradiction:
Improveboundary tracking accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent divides the complex tracking system into two specialized pipelines with distinct functions. The first pipeline handles real-time video processing and feature extraction using a simple set of trackable points, while the second pipeline handles the complex calculations for determining precise location parameters. This segmentation reduces the complexity burden on any single component while achieving high overall precision.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent uses an intermediary data exchange mechanism between the two pipelines. The first pipeline transfers feature point data to the second pipeline, which then returns calculated location parameters. This intermediary approach allows the system to achieve accurate boundary tracking without requiring either pipeline to be overly complex, as each pipeline focuses on its specific task.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Measurement precision

If location parameters are recalculated for every frame and time interval to maintain tracking accuracy, then measurement precision is improved, but loss of time increases

Engineering Contradiction:
Improvetracking accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary action by having the second pipeline pre-calculate precise location parameters based on feature points extracted by the first pipeline. These pre-calculated parameters are then transferred back to the first pipeline for augmentation, eliminating the need for the first pipeline to perform time-consuming recalculations for every frame while maintaining high tracking accuracy.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP4068204B1Method and system for augmenting point of interest in augmented-reality video
Publication Date: 2025.07.02 REVIEVE OY
  • EP4068204B1 patent drawingFigure 1~2B
  • EP4068204B1 patent drawingFigure 2C~3
  • EP4068204B1 patent drawingFigure 4

AI summary

A method for augmenting point of interest (202) in AR video. The method includes: extracting first frame (300) of AR video depicting object being tracked (200) including point of interest (202), by first pipeline; locating, in first frame (300) set of feature points forming boundary (206) of object; transferring frame (300) with determined set of feature points to second pipeline; and by second pipeline determining coordinate system (210) of set of feature points; calculating first location parameters of feature points; selecting first and second reference points (216) from feature points, wherein first location parameters of point of interest (202) are defined by first location parameters of first and second reference points (216); transmitting first location parameters of point of interest (202) from second pipeline to first pipeline; and augmenting in first pipeline AR video with first location parameters of point of interest (202) received from second pipeline.