3D Scene Reconstruction for Rapid Scalable Vector Drawings

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for achieving three-dimensional spatial and semantic understanding of a location, such as for home renovations or insurance claims, rely on manual measurements and hard-to-acquire architectural drawings, which are time-consuming and require coordination with multiple parties.

Innovation Solution

A system that generates isometric and orthographic vector drawings based on a three-dimensional representation of a physical scene using scalable vector graphics (SVG), allowing for real-time rendering and infinite scaling without blurring, utilizing a trained machine learning model to process image and sensor data to identify and annotate surfaces and contents.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual measurements and traditional architectural drawing methods are used, then measurement accuracy and drawing precision are maintained, but the time required for data collection and drawing generation becomes excessively long

Engineering Contradiction:
Improvespatial understanding accuracyVSAvoidtime for manual measurements and drawing generation
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces manual mechanical measurement systems with an automated computer vision system. The machine learning model processes images and videos captured by cameras or smartphones to automatically generate 3D representations and architectural drawings, eliminating the need for manual tape measures, laser distance meters, and manual sketching operations while maintaining measurement precision

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

Solution Approach 2:

The system creates a virtual 3D copy of the physical space by processing multiple images and videos. This digital twin captures the spatial relationships, dimensions, and semantic information of the environment, allowing accurate architectural drawings to be generated from the virtual representation without requiring physical measurement tools or onsite drafting

Inventive Principle:
Principle #26Copying

2Manufacturing precision

If traditional architectural drawing methods are used, then drawing accuracy is maintained, but the complexity of coordinating multiple parties with competing schedules increases

Engineering Contradiction:
Improvearchitectural drawing accuracyVSAvoidcoordination complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The system performs self-service by automatically capturing images and videos, processing them through machine learning models, generating 3D representations, and creating architectural drawings without requiring coordination between surveyors, drafters, or other professionals. The automated system independently completes the entire workflow from data collection to drawing generation

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent merges multiple functions that were previously performed by separate parties into a single integrated system. The machine learning model simultaneously performs image processing, 3D reconstruction, dimension extraction, and drawing generation, eliminating the need for coordination between surveyors, drafters, and other specialists

Inventive Principle:
Principle #5Merging (Combining)

3Ease of operation

If raster image formats are used for architectural drawings, then the drawings can be displayed on standard devices, but the ability to zoom without blurring is limited

Engineering Contradiction:
Improvedrawings display compatibilityVSAvoidzoom clarity
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The system changes the fundamental parameter of the drawing format from raster (pixel-based) to vector (mathematical equation-based). Vector drawings use mathematical descriptions of lines, shapes, and text that can be scaled to any size without loss of quality, while maintaining compatibility with standard viewing devices through SVG or PDF formats

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS20250316022A1Generating vector drawings based on a three dimensional representation of a physical scene at a location
Publication Date: 2025.10.09 YEMBO INC
  • US20250316022A1 patent drawing
  • US20250316022A1 patent drawing
  • US20250316022A1 patent drawing

AI summary

The described systems and methods are configured generate a 3D virtual representation of a physical scene at a location, and output isometric and/or orthographic vector drawings based on the 3D virtual representation. The vector drawings are generated by rendering views of the 3D virtual representation in a scalable vector graphics (SVG) format, so that the views can be zoomed without blurring or other decreases in image viewability. Further, sub-rooms, tags, labels, and/or dimensions may be added to the output. Rather than taking hours to generate drawings, the described systems and methods enable generation in a few milliseconds, among other advantages.