3D Data Point Assignment Using 2D Polygons and Depth Classes

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing 3D object recognition methods based on machine learning are computationally intensive and require significant manual effort for data annotation, making them inefficient, especially for large datasets.

Innovation Solution

A method for assigning data points from three-dimensional data to objects using two-dimensional surrounding polygons and depth classes, employing a projection matrix and the Hungarian algorithm to form clusters efficiently, reducing computational requirements.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If machine learning-based 3D object recognition methods are used, then recognition accuracy is improved, but computational complexity and processing time increase significantly

Engineering Contradiction:
Improve3D object recognition accuracyVSAvoidcomputational complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the 3D point cloud data into multiple two-dimensional projections (front, back, left, right views). Each projection is processed independently to generate 2D surrounding polygons, which are then integrated to form the final 3D object recognition result. This segmentation reduces the computational complexity of processing large 3D datasets while maintaining recognition accuracy.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transforms the 3D object recognition problem into multiple 2D problems by projecting 3D point cloud data onto two-dimensional planes. The projection matrix converts 3D coordinates (x, y, z) into 2D coordinates (x', y'), allowing the use of computationally efficient 2D image processing algorithms while still achieving accurate 3D object recognition through integration of multiple views.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Measurement precision

If manual data annotation is performed for training datasets, then data quality and labeling accuracy are improved, but time consumption and costs increase

Engineering Contradiction:
Improvedata labeling accuracyVSAvoidannotation time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent enables the system to automatically generate 2D surrounding polygons and assign depth classes to objects without requiring manual annotation. The projection-based method self-generates training data by processing 3D point cloud information through mathematical projections, eliminating the need for time-consuming manual labeling while maintaining high data quality.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent performs preliminary processing of 3D point cloud data by generating projected 2D views and automatically creating surrounding polygons before the actual object recognition task. This preliminary action prepares the data in advance with accurate geometric information, reducing the need for subsequent manual annotation and speeding up the overall processing pipeline.

Inventive Principle:
Principle #10Preliminary action

3Ease of operation

If traditional clustering methods are used for assigning data points to objects, then simplicity is maintained, but accuracy decreases due to overlapping polygons and depth ambiguities

Engineering Contradiction:
Improvemethod simplicityVSAvoiddata point assignment accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The patent introduces depth classes as an additional dimension to resolve ambiguities in 2D projections. By classifying data points into different depth levels based on their z-coordinate values, the method accurately assigns points to objects even when 2D surrounding polygons overlap, maintaining simplicity while improving precision.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The patent introduces depth classes as an intermediary mechanism to mediate between 2D projected data and 3D object assignment. The depth class acts as a mediator that resolves conflicts when multiple objects occupy the same 2D space, enabling accurate data point assignment without complex computational algorithms.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP4693182A1Method for assigning data points to one of a plurality of objects
Publication Date: 2026.02.11 DSPACE SE & CO KG
  • EP4693182A1 patent drawingFigure 1~2
  • EP4693182A1 patent drawingFigure 3~4
  • EP4693182A1 patent drawingFigure 5a~5b

AI summary

The invention relates to a computer-implemented method for assigning data points from first data (26) to each of several objects (14, 14'), wherein the first data (26) comprise at least three spatial dimensions, and the method comprises the following step: - forming groups of projected data points (18, 18') assigned to the respective objects (14, 14') taking into account received information data from two-dimensional surrounding polygons (10, 10') and from a depth class (20, 20') of the data point by means of an assignment method such that, in the case of overlapping polygons (10, 10') where projected data points (18, 18') are located within more than one two-dimensional surrounding polygon (10, 10'), the groups are formed such that no depth class (20, 20') is simultaneously assigned to several of the overlapping polygons (10, 10, 10'). 10') is assigned.Furthermore, the invention relates to a data processing device comprising means for carrying out the above method, as well as a computer program product comprising instructions that, when executed by a computer, cause the computer to carry out the above method. The invention also relates to a computer-readable data carrier on which the above computer program product is stored.