Point Cloud Storage Using Half-Precision for Real-Time Scanning

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods to reduce memory requirements for point cloud data either degrade data quality or require significant computational resources, making them unsuitable for high-precision and real-time applications.

Innovation Solution

Convert point cloud data of valid points from single-precision floating-point format to half-precision floating-point format, and store it in this format, while determining valid points based on a preset matching condition.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If point cloud data is stored in single-precision floating-point format, then data quality and precision are maintained, but memory requirements increase

Engineering Contradiction:
Improvedata precisionVSAvoidmemory requirements
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent changes the data precision parameter from single-precision (32-bit) to half-precision (16-bit) floating-point format. This parameter change reduces the memory footprint by 50% while maintaining sufficient precision for scanning applications, directly resolving the contradiction between data precision and memory requirements

Inventive Principle:
Principle #35Parameter changes

2Quantity of substance

If data compression is applied to reduce memory requirements, then memory usage decreases, but computational resources increase

Engineering Contradiction:
Improvememory usageVSAvoidcomputational resources
Core Design Contradiction:
Quantity of substanceVSUse of energy by moving object

Solution Approach 1:

The patent uses a simpler, more efficient data representation format (half-precision floating-point) that requires no complex compression algorithms. This approach achieves memory reduction through format conversion rather than computational compression, eliminating the trade-off between memory savings and computational overhead

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

3Quantity of substance

If resolution is lowered to reduce memory requirements, then memory requirements decrease, but data quality degrades

Engineering Contradiction:
Improvememory requirementsVSAvoiddata quality
Core Design Contradiction:
Quantity of substanceVSMeasurement precision

Solution Approach 1:

Instead of reducing the number of points (resolution), the patent changes the data format precision parameter from 32-bit to 16-bit floating-point. This maintains the same point cloud resolution and data quality while achieving memory reduction through more efficient data encoding

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP4679269A1Storage method and apparatus for point cloud data, and device and medium
Publication Date: 2026.01.14 SHINING 3D TECH CO LTD
  • EP4679269A1 patent drawingFigure 1
  • EP4679269A1 patent drawingFigure 2
  • EP4679269A1 patent drawingFigure 3

AI summary

The present disclosure relates to relates to a method, an apparatus, a device, and a medium for storing point cloud data. The method includes: acquiring point cloud data of each point in a single frame image; determining valid points satisfying a preset matching condition from all points in the single frame image based on the point cloud data and a preset matching condition; converting the point cloud data of the valid points from a single-precision floating-point format to a half-precision floating-point format and storing the point cloud data with the half-precision floating-point format. According to the technical solution of the present disclosure, the memory requirements for scanning tasks can be reduced while retaining data details and ensuring real-time performance.