In-Tree Point Cloud Quantization for Spatially Adaptive Compression

Solution Overview

Problem

Point clouds captured by sensors like LIDAR systems and 3-D cameras contain large amounts of data, making storage and transmission costly and time-consuming, which limits their use in real-time applications.

Innovation Solution

A method involving a tree data structure is used to encode and compress point cloud data, where different quantization step sizes are applied to subsets of points based on their spatial isolation, reducing the data size and enabling efficient storage and transmission.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If point cloud data is stored and transmitted in full precision, then measurement precision is maintained, but storage cost and transmission time increase significantly

Engineering Contradiction:

Improvepoint cloud position precisionVSAvoidtransmission time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent applies different quantization step sizes to different subsets of points based on their spatial isolation characteristics. Spatially isolated points use a first quantization step size while non-isolated points use a second quantization step size, allowing optimized precision for each region rather than uniform precision throughout the entire point cloud

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent changes the quantization parameter (step size) based on the spatial distribution characteristics of points. By detecting spatial isolation and adjusting the quantization step size accordingly, the system adapts the precision level to match the actual information content needed for different regions of the point cloud

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If point cloud data is stored and transmitted in full precision, then measurement precision is maintained, but storage cost increases significantly

Engineering Contradiction:

Improvepoint cloud position precisionVSAvoidstorage requirement

Core Design Contradiction:

Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent applies different quantization step sizes to different subsets of points based on their spatial isolation characteristics. Spatially isolated points use a first quantization step size while non-isolated points use a second quantization step size, allowing optimized precision for each region rather than uniform precision throughout the entire point cloud

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent changes the quantization parameter (step size) based on the spatial distribution characteristics of points. By detecting spatial isolation and adjusting the quantization step size accordingly, the system adapts the precision level to match the actual information content needed for different regions of the point cloud

Inventive Principle:

Principle #35Parameter changes

3Device complexity

If uniform quantization is applied to all points, then processing complexity is reduced, but data compression efficiency decreases

Engineering Contradiction:

Improvequantization processing complexityVSAvoiddata compression efficiency

Core Design Contradiction:

Device complexityVSProductivity

Solution Approach 1:

The patent segments the point cloud into different subsets based on spatial isolation characteristics. By dividing points into spatially isolated and non-isolated groups, the system can apply different quantization strategies to each segment, improving overall compression efficiency while maintaining manageable processing complexity through systematic classification

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies different quantization step sizes to different subsets of points based on their spatial isolation characteristics. Spatially isolated points use a first quantization step size while non-isolated points use a second quantization step size, allowing optimized precision for each region rather than uniform precision throughout the entire point cloud

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS20250274607A1In-tree geometry quantization of point clouds

Publication Date: 2025.08.28 APPLE INC

AI summary

An example device includes one or more processors, and memory storing instructions that when executed by the processors, cause the processors to receive points that represent a point cloud in three-dimensional space, and generate a data structure representing the point cloud. Generating the data structure includes encoding a position of each point in each dimension as a sequence of bits according to a tree data structure; partitioning each of the sequences into two or more portions according to a scaling depth; determining that a subset of the points is spatially isolated from a remainder of the points; quantizing each of the portions associated with the subset of the points according to a first quantization step size; quantizing each of the portions associated with the remainder of the points according to a second quantization step size; and including the quantized portions in the data structure.