3D Model Generation from Irregular Sketches via Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current sketch programs require significant effort and adaptation to convert 2D sketches into 3D models, and if corrections are needed, designers must start over from scratch, limiting design concept development through simulations.
Innovation Solution
A method and system using machine learning models to automatically generate and correct 3D models from irregular-shaped 2D sketches, allowing users to input additional strokes for real-time updates and recommendations, enabling efficient design simulation and correction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If traditional 3D conversion technology is used to convert 2D sketch images into 3D images, then 3D models can be generated, but it requires extra effort and adaptation period for designers and has limitations on drawing programs
Solution Approach 1:
The system enables self-service by automatically generating 3D models from 2D sketches without requiring designer intervention in the conversion process. The AI model autonomously performs the 3D conversion, eliminating the need for designers to learn specialized 3D drawing techniques or use complex conversion tools.
Solution Approach 2:
The patent replaces the mechanical/manual process of 3D model creation with an automated AI-based system. Instead of designers manually creating 3D models or using traditional conversion tools, an neural network model automatically performs the conversion from 2D sketches to 3D models, substituting human effort with intelligent automation.
2Adaptability or versatility
If corrections are needed for the converted 3D model, then the 2D sketch image has to be sketched from the beginning, but this causes difficulties in developing design concepts through simulations
Solution Approach 1:
The system introduces dynamics by allowing iterative modifications where corrections to the 3D model automatically trigger regeneration based on updated 2D sketches. This dynamic process enables designers to refine design concepts through multiple simulations without starting from scratch, making the design exploration flexible and adaptive.
Solution Approach 2:
The patent implements feedback by automatically regenerating 3D models when corrections are made to 2D sketches. The system provides real-time feedback to designers by updating the 3D model based on modified sketches, enabling continuous refinement of design concepts without manual rework or time loss.
3Ease of manufacture
If traditional sketch programs are used, then 2D sketch images can be drawn, but there are limitations on programs for drawing 2D sketch images and extra effort for conversion
Solution Approach 1:
The patent extracts the complex 3D conversion technology from the user's workflow by implementing it as an automated background process. The complexity of 3D model generation is separated from the simple 2D sketching interface, allowing designers to use straightforward drawing programs while the conversion complexity is handled autonomously by the system.
Data Source
AI summary
Provided is a method for 3D modeling based on an irregular-shaped sketch, in which the method is executed by one or more processors, and includes receiving 2D sketch data of a target object, inputting the 2D sketch data into a 3D model generation model to generate a 3D model of the target object, and displaying the generated 3D model on a display.