Automated Building Video Generation From Floor Plans
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for representing and utilizing building interior information, such as floor plans, are inefficient and often inaccurate, making it difficult to effectively capture, visualize, and navigate building interiors remotely, especially for architectural analysis, real estate, and contracting purposes.
Innovation Solution
The development of a system that automatically generates building videos using acquired images and floor plans, employing machine learning models to analyze and synthesize visual data, allowing for the creation of panoramic and navigable virtual tours without the need for depth sensors, and enabling the identification of building attributes and generation of accompanying narratives.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If floor plans are used to represent building interiors, then some layout information is provided, but they are difficult to construct, maintain, accurately scale, and visualize
Solution Approach 1:
The patent creates virtual copies of building interiors through automated video generation from floor plans. The system generates photorealistic video representations that replicate the visual appearance of building interiors without requiring physical measurement devices or complex manual floor plan construction, thereby preserving building interior information while eliminating construction and maintenance complexity
Solution Approach 2:
The patent replaces manual floor plan construction and visualization methods with an automated machine learning-based video generation system. This substitution eliminates the need for manual scaling, drawing, and maintenance of floor plans, while providing more intuitive and accurate building interior representations through generated videos
2Loss of information
If textual descriptions of buildings are used, then some building information is provided, but they are often inaccurate and incomplete
Solution Approach 1:
The patent generates accurate visual copies of building interiors and attributes through automated video synthesis. By creating photorealistic representations from floor plans and images, the system captures complete and accurate building attributes including spatial relationships, materials, and features that textual descriptions often miss or misrepresent
Solution Approach 2:
The patent introduces an automated machine learning-based video generation system as an intermediary between floor plan data and building representation. This intermediary processes structured floor plan information and images to generate accurate, complete, and visually intuitive building descriptions, eliminating the inaccuracies of manual textual descriptions
3Loss of information
If physical travel to buildings is required for interior inspection, then complete building interior information is obtained, but time and resource consumption increase
Solution Approach 1:
The patent creates virtual video copies of building interiors that can be inspected remotely. These generated videos provide complete building interior detail information including spatial layouts, materials, and features, allowing users to conduct thorough inspections without physically traveling to the building, thereby eliminating time and resource consumption associated with on-site visits
4Productivity
If automated video generation from floor plans is implemented, then computational resources and time are reduced, but the system complexity increases
Solution Approach 1:
The patent implements a self-service automated video generation system that processes floor plans and images to create building videos without manual intervention. The machine learning models automatically perform video synthesis, narration generation, and quality assessment, eliminating the need for manual video production resources while achieving high productivity through autonomous operation
Data Source
AI summary
Techniques are described for using computing devices to perform automated operations for automatically generating videos and associated information about a building interior using other visual data about the building interior, as well as presenting the generated videos and associated information in various manners. In some situations, the generation is based at least in part on user input provided via user interactions with a displayed floor plan of the building, such as to select one or more rooms or other areas for which to include visual data in the video, and/or to select one or more building objects and/or other building structural elements and/or other building attributes for which to include visual data in the video. The techniques may further include determining and using information about such building attributes of the building from automated analysis of building information that includes floor plans and acquired building images.