Multi-dimensional Style Transfer for 3D Object Models

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current design processes are inefficient and resource-intensive due to their iterative nature, lacking tools to reduce the number of iterations and increase efficiency in capturing appealing product designs, particularly in applying styles and shapes between objects.

Innovation Solution

A computer-implemented intelligent design platform utilizing multi-dimensional style transfer through machine learning models, such as MeshCNN and PointNet, to segment and merge styles between source and target objects, enabling the application of a source object's style or shape to a target object by calculating compatibility scores and selectively replacing segments.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional iterative design processes are used to capture appealing designs, then design flexibility and creativity are maintained, but resource consumption increases and productivity decreases

Engineering Contradiction:

Improvedesign iteration efficiencyVSAvoidprocessor and memory resource consumption

Core Design Contradiction:

ProductivityVSUse of energy by stationary object

Solution Approach 1:

The system creates digital copies of object models and applies style transfer algorithms to generate variations without requiring manual redrawing or complete redesign. The style transfer process copies stylistic features from source objects to target objects, enabling efficient exploration of design variations while maintaining original model structures.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent replaces manual iterative design processes with automated machine learning-based style transfer systems. Instead of designers manually adjusting models through multiple iterations, the system automatically transfers styles between objects using neural networks, significantly reducing computational resource requirements and time consumption.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Manufacturing precision

If manual iterative design processes are used, then design quality can be refined through multiple iterations, but the time required for design increases

Engineering Contradiction:

Improvedesign qualityVSAvoiddesign iteration time

Core Design Contradiction:

Manufacturing precisionVSLoss of time

Solution Approach 1:

The system performs preliminary style extraction and feature analysis automatically, preparing style representations and compatibility assessments before the actual design iteration. This preliminary processing enables rapid style transfer and reduces the time needed for manual iteration while maintaining design quality.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system incorporates compatibility scoring mechanisms that provide immediate feedback on style suitability between objects. By calculating compatibility scores and presenting them to users for selection, the system enables rapid iteration with guided feedback, reducing the time required to achieve desired design quality compared to blind manual iteration.

Inventive Principle:

Principle #23Feedback

3Productivity

If style transfer is applied between objects, then design efficiency increases and resource demands are reduced, but the complexity of the system increases

Engineering Contradiction:

Improvedesign efficiencyVSAvoidsystem complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The system segments objects into distinct style components and content components, allowing independent processing and transfer of stylistic features without affecting the underlying object structure. This segmentation enables complex style transfer operations to be broken down into manageable steps, reducing overall system complexity while maintaining high productivity.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces compatibility scoring as an intermediary mechanism that mediates between source objects and target objects during style transfer. This intermediary layer simplifies the complex task of style matching by providing automated compatibility assessments, making the system more manageable and less complex while maintaining efficient design productivity.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS11704802B2Multi-dimensional model merge for style transfer

Publication Date: 2023.07.18 ACCENTURE GLOBAL SOLUTIONS LTD

AI summary

Implementations are directed to receiving a target object model representative of a target object, receiving a source object model representative of a source object, defining a set of target segments and a set of source segments using a segmentation machine learning (ML) model, for each target segment and source segment pair in a set of target segment and source segment pairs, generating a compatibility score representing a degree of similarity between a target segment and a source segment, the compatibility score calculated based on global feature representations of each of the target segment and the source segment, each global feature representation determined from a ML model, selecting a source segment for style transfer based on compatibility scores, and merging the source segment into the target object model to replace a respective target segment within the target object model and providing a stylized target object model.