3D Positioning Model Clustering for Mobile AR Scene Localization

Solution Overview

Problem

Conventional image-based positioning methods require high computing power and network connectivity, limiting their application on mobile devices and real-time Augmented Reality (AR) applications due to the large number of 3D points and descriptors in the positioning model.

Innovation Solution

A method to optimize the positioning model by clustering 3D points and their descriptors based on distance relationships, reducing redundancy and enabling efficient positioning on mobile devices.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a positioning model with a large number of 3D points and descriptors is used to improve positioning accuracy, then positioning precision is improved, but computing power requirements and device complexity increase

Engineering Contradiction:

Improvepositioning accuracyVSAvoidcomputing power requirements

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts and removes redundant 3D points and descriptors from the positioning model through clustering analysis. By identifying and eliminating duplicate or highly similar feature points, the model maintains positioning accuracy while significantly reducing the total number of points and descriptors that need to be processed, thereby lowering computing power requirements.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies different processing treatments to different regions of the positioning model based on local characteristics. Through clustering, it identifies regions with high point density and redundancy, applying optimization specifically to these local areas while preserving important distinctive features elsewhere, thus achieving efficiency improvement without compromising overall accuracy.

Inventive Principle:

Principle #3Local quality

2Measurement precision

If a positioning model with a large number of 3D points and descriptors is used to improve positioning accuracy, then positioning precision is improved, but processing speed decreases

Engineering Contradiction:

Improvepositioning accuracyVSAvoidprocessing speed

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent extracts and removes redundant 3D points and descriptors from the positioning model through clustering analysis. By identifying and eliminating duplicate or highly similar feature points, the model maintains positioning accuracy while significantly reducing the total number of points and descriptors that need to be processed, thereby lowering computing power requirements.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent performs preliminary optimization of the positioning model by pre-removing redundant 3D points and descriptors before the actual positioning operation. This preliminary action of clustering and filtering is done once during model preparation, so that during real-time positioning, the system only needs to process the already-optimized subset of points, dramatically improving processing speed.

Inventive Principle:

Principle #10Preliminary action

3Reliability

If conventional positioning methods are used to ensure comprehensive scene coverage, then positioning reliability is improved, but applicability to mobile devices and real-time AR applications deteriorates

Engineering Contradiction:

Improvepositioning reliabilityVSAvoidapplicability to mobile devices

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent extracts and removes redundant 3D points and descriptors from the positioning model through clustering analysis. By identifying and eliminating duplicate or highly similar feature points, the model maintains positioning accuracy while significantly reducing the total number of points and descriptors that need to be processed, thereby lowering computing power requirements.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent changes the parameter of the positioning model by reducing the number of 3D points and descriptors through clustering-based optimization. This parameter change transforms the model from a computationally intensive format suitable for servers to a streamlined format that can be efficiently processed on mobile devices with limited computing resources, while maintaining positioning reliability.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS12494028B2Positioning model optimization method, positioning method, positioning device, and storage medium

Publication Date: 2025.12.09 BYTEDANCE INC

AI summary

A positioning model optimization method, an image-based positioning method and positioning device, and a computer-readable storage medium are provided. The positioning model optimization method includes: inputting a positioning model for a scene, the positioning model including a three-dimensional (3D) point cloud and a plurality of descriptors corresponding to each 3D point in the 3D point cloud; determining, for each 3D point in the 3D point cloud, a plurality of neighboring points of the 3D point, and if a distance relationship between each of the plurality of neighboring points and the 3D point is smaller than a predetermined threshold, outputting the 3D point and the plurality of descriptors corresponding to the 3D point to an optimized positioning model for the scene; and outputting the optimized positioning model for the scene.