3D Voxel Data Encoding Device Structure Attribute Separation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing encoding methods for three-dimensional voxel data used in 3D printing are inefficient, leading to potential mosquito noise during decoding, where attribute values may appear at incorrect lattice positions, causing manufacturing errors.
Innovation Solution
An encoding device that separates structure and attribute information, using context-based encoding for both, where structure information is encoded based on the presence/absence of voxels and attribute values are encoded using context information about surrounding voxels, to improve encoding efficiency and reduce noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If encoding methods for two-dimensional images are applied to three-dimensional voxel data, then encoding can be performed, but encoding efficiency is insufficient due to the sparseness characteristic of 3D voxel data
Solution Approach 1:
The patent segments the encoding process into two distinct parts: structure information encoding (presence/absence of voxels) and attribute information encoding (color and other properties). This segmentation allows each part to be optimized separately, with structure encoding handling the sparseness characteristic and attribute encoding handling the color data, thereby resolving the contradiction between encoding efficiency and data accuracy.
Solution Approach 2:
The patent applies different encoding strategies to different parts of the voxel data based on their local characteristics. Structure information is encoded using methods suitable for sparse data, while attribute information is encoded using methods optimized for color data. This local quality approach ensures that each encoding process is optimized for its specific data type, improving overall encoding efficiency without sacrificing accuracy.
2Productivity
If context information is used for encoding structure information, then encoding efficiency improves, but encoding complexity increases
Solution Approach 1:
The patent divides the encoding process into separate stages: structure information encoding and attribute information encoding. By segmenting the process, context information is only used in the structure encoding stage where it is most beneficial, rather than throughout the entire encoding process. This reduces overall complexity while maintaining encoding efficiency benefits.
3Quantity of substance
If voxel data is compressed to reduce data amount, then storage efficiency improves, but manufacturing precision deteriorates due to mosquito noise and incorrect attribute values
Solution Approach 1:
The patent separates structure and attribute information into distinct encoding streams. Structure information is encoded with high precision to maintain geometric accuracy, while attribute information is encoded separately with error correction. This segmentation allows compression to be applied more aggressively to attribute data without compromising the precision of structure data, thereby maintaining manufacturing precision while reducing overall data amount.
Solution Approach 2:
The patent employs feedback mechanisms in the decoding process where the decoded structure information is used to verify and correct attribute information. This feedback loop ensures that even if compression introduces errors in attribute data, the structure information serves as a reference to maintain manufacturing precision and prevent incorrect attribute values from affecting the final model accuracy.
Data Source
AI summary
An encoding device includes an extracting unit that extracts, from voxel data expressing a three-dimensional shape by a collection of voxels arranged at lattice positions in a space, (i) structure information indicating presence/absence of a voxel at each lattice position and (ii) attribute information including an attribute value of each voxel; a structure encoding unit that encodes the structure information by determining a structure code corresponding to the structure information on a structure element having at least lattice position according to first context information indicating presence/absence of voxel at at least one predetermined lattice position around the structure element; and an attribute encoding unit that encodes an attribute value of a voxel of interest into an attribute code using second context information indicating presence/absence of a voxel at at least one predetermined lattice position around the voxel of interest.