RAHT Point Cloud Decoding for Outlier-Resistant Attribute Prediction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional point cloud decoding technologies are adversely affected by outliers during smoothing processes, leading to inefficiencies in encoding attribute information.

Innovation Solution

The implementation of a point cloud decoding device that uses clipping and scaling factors for attribute value intra-prediction in subnodes, along with RAHT units for smoothing and inter prediction of AC and DC coefficients, to enhance encoding efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If smoothing is performed using conventional methods on AC coefficients, then the decoding process can be completed, but the smoothing process is significantly affected by outliers leading to degraded encoding efficiency

Engineering Contradiction:

Improveencoding efficiencyVSAvoidoutlier impact

Core Design Contradiction:

ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The patent converts the harmful effect of outliers into a beneficial filtering process by identifying outlier points through statistical analysis (comparing attribute values against mean and standard deviation) and excluding them from the smoothing calculation. This transforms the problem of outlier interference into an opportunity to improve smoothing quality through selective processing.

Inventive Principle:

Principle #22Blessing in disguise (Convert harm into benefit)

Solution Approach 2:

The patent introduces new parameters for outlier detection (threshold values based on standard deviation multiples) and modifies the smoothing process by conditionally including or excluding points based on these parameters. This allows dynamic adjustment of smoothing behavior to adapt to different data conditions and minimize outlier impact.

Inventive Principle:

Principle #35Parameter changes

2Productivity

If clipping is applied to intra-predicted attribute values in RAHT units, then encoding efficiency is improved, but the device complexity increases

Engineering Contradiction:

Improveencoding efficiencyVSAvoiddecoding process complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent applies clipping operations during the intra-prediction stage in the RAHT unit, performing the amplitude limitation beforehand rather than during final reconstruction. This preliminary action ensures that subsequent processing stages receive already-optimized data, improving overall encoding efficiency while concentrating the computational complexity in a specific, manageable location.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The clipping operation is applied locally to specific attribute values based on their predicted ranges, rather than uniformly to all data. This localized processing approach optimizes encoding efficiency for regions where clipping is beneficial while avoiding unnecessary processing elsewhere, balancing complexity and performance.

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS20260105645A1Point cloud decoding device, point cloud decoding method, and program

Publication Date: 2026.04.16 KDDI CORP

AI summary

A point cloud decoding device 200 according to the present invention includes: an RAHT unit 2080 configured to, in inter prediction of an AC coefficient for each node, apply a scaling factor to a predicted value of the AC coefficient or a predicted value of an attribute value. According to the present invention, it is possible to provide a point cloud decoding device, a point cloud decoding method, and a program capable of improving encoding efficiency in encoding attribute information.