AI 2D-to-3D Asset Generation With Automated Image Validation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Online retailers face challenges in providing customers with 3D representations of products, as generating 3D assets typically requires manual effort by skilled artists, limiting scalability and efficiency.

Innovation Solution

An automated 3D asset generation framework using AI pipelines processes 2D images to create 3D assets, employing deep convolutional neural networks, image segmentation, and super-resolution models to enhance and validate image quality, enabling efficient and accurate conversion of 2D images into 3D models for virtual environments.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If manual creation by experienced 3D artists is used, then high-quality 3D assets are produced, but productivity is low and costs are high

Engineering Contradiction:
Improve3D asset qualityVSAvoid3D asset generation speed
Core Design Contradiction:
Manufacturing precisionVSProductivity

Solution Approach 1:

The patent replaces manual mechanical 3D modeling work with an automated AI-based system that uses computer vision and machine learning algorithms to generate 3D assets from 2D images, eliminating the need for manual artist intervention while maintaining quality standards

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system creates accurate 3D copies of physical products by capturing 2D images and using AI algorithms to reconstruct three-dimensional models, enabling digital representations that faithfully replicate the original objects without manual modeling

Inventive Principle:
Principle #26Copying

2Manufacturing precision

If manual creation by experienced 3D artists is used, then high-quality 3D assets are produced, but the process is not scalable

Engineering Contradiction:
Improve3D asset qualityVSAvoidscalability
Core Design Contradiction:
Manufacturing precisionVSAdaptability or versatility

Solution Approach 1:

The patent replaces manual mechanical 3D modeling work with an automated AI-based system that uses computer vision and machine learning algorithms to generate 3D assets from 2D images, eliminating the need for manual artist intervention while maintaining quality standards

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system is designed to handle multiple types of products and 2D image formats through a universal AI pipeline that can process various inputs and generate corresponding 3D assets across different categories, enabling scalable application across diverse product lines

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Productivity

If automated AI pipelines are used, then productivity and scalability are improved, but image processing complexity increases

Engineering Contradiction:
Improve3D asset generation speedVSAvoidimage processing complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent divides the complex image processing task into distinct sequential stages including image input, 3D model generation, and validation phases, with each stage handled by specialized AI components that process specific aspects of the transformation independently

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system introduces intermediate processing steps and validation layers that mediate between the input 2D images and final 3D outputs, using AI models to bridge the transformation while maintaining quality control and reducing overall system complexity through modular architecture

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS20250245913A1Automated 3D asset generation framework
Publication Date: 2025.07.31 WALMART APOLLO LLC
  • US20250245913A1 patent drawing
  • US20250245913A1 patent drawing
  • US20250245913A1 patent drawing

AI summary

A system including a processor and a non-transitory computer-readable media storing computing instructions that, when executed on the processor, cause the processor to perform certain operations. The operations can include identifying a 2D silo image of a geometric item based on a probability value exceeding a predetermined probability threshold. The operations also can include segmenting artifacts from the 2D silo image to isolate first pixels of a border of the geometric item. The operations further can include trimming second pixels along the border of the geometric item. The operations also can include performing an aspect ratio validation on the 2D silo image to validate that the 2D silo image corresponds to a shape of the geometric item. The operations additionally can include auto-validating that a visual resolution level of the 2D silo image falls within a predetermined acceptance rate. The operations further can include generating a 3D view image from the 2D silo image of the geometric item enabled for use in virtual environments when the visual resolution level falls within the predetermined acceptance rate. Other embodiments are described.