3D Model Co-segmentation via Position-Constrained Affinity Propagation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional affinity propagation models for co-segmentation of three-dimensional model sets lack consideration for position relation information between neighboring super patches, resulting in low consistency between co-segmentation outcomes.
Innovation Solution
A consistent affinity propagation model is developed, constrained by predefined position information between super patches, which is converted into a consistent convergence affinity propagation model to improve clustering and generate more consistent co-segmentation outcomes by incorporating position relations and robust feature vectors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional affinity propagation model is used for co-segmentation, then the clustering process is simple, but the consistency between co-segmentation outcomes deteriorates due to lack of position relation information
Solution Approach 1:
The patent segments the affinity propagation model into two distinct components: a conventional affinity propagation model for basic clustering and a consistent affinity propagation model that incorporates position relation information. This segmentation allows the system to maintain the simplicity of conventional AP while adding positional constraints to improve consistency, resolving the contradiction between model simplicity and outcome reliability.
Solution Approach 2:
The patent introduces position relation information as an intermediary element that mediates between super patches and cluster centers. This intermediary carries spatial relationship data that enhances the consistency of co-segmentation outcomes without fundamentally altering the affinity propagation mechanism, thus improving reliability while controlling complexity.
2Reliability
If position relation information is incorporated into affinity propagation model, then the consistency of co-segmentation outcomes is improved, but the model complexity increases
Solution Approach 1:
The patent applies local quality by incorporating position relation information specifically where needed - in the consistent affinity propagation model - rather than uniformly across all clustering operations. This allows the system to enhance consistency in critical areas while maintaining the simplicity of the conventional AP model for general-purpose clustering, thus managing overall complexity.
Solution Approach 2:
The patent implements partial action by selectively applying position relation constraints only to certain aspects of the affinity propagation process through the consistent AP model, rather than completely transforming the entire clustering framework. This partial incorporation improves consistency while avoiding the full complexity burden of a complete model redesign.
3Measurement precision
If robust feature vectors and association relations are extracted and integrated, then the accuracy of co-segmentation results is improved, but the processing time and computational complexity increase
Solution Approach 1:
The patent applies preliminary action by pre-establishing the consistent convergence affinity propagation model with integrated position relation information and robust feature vectors before actual co-segmentation operations. This preprocessing step, while time-consuming, is performed once to enable faster and more accurate clustering operations in subsequent executions, thus reducing the time penalty for repeated operations.
Solution Approach 2:
The patent changes key parameters of the affinity propagation model by integrating position relation information and robust feature vectors, transforming the conventional AP model into a consistent convergence AP model. These parameter changes enhance measurement precision by incorporating additional dimensional information, while the one-time nature of the transformation minimizes recurring time losses.
Data Source
AI summary
Disclosed is a co-segmentation method and apparatus for a three-dimensional model set, which includes: obtaining a super patch set for the three-dimensional model set which includes at least two three-dimensional models, each of the three-dimensional models including at least two super patches; obtaining a consistent affinity propagation model according to a first predefined condition and a conventional affinity propagation model, the consistent affinity propagation model being constraint by the first predefined condition which is position information for at least two super patches that are in the super patch set and belong to a common three-dimensional model set; converting the consistent affinity propagation model into a consistent convergence affinity propagation model; clustering the super patch set through the consistent convergence affinity propagation model to generate a co-segmentation outcome for the three-dimensional model set. The disclosed three-dimensional model set co-segmentation method and apparatus improves consistency between three-dimensional model set co-segmentation outcomes.