Automated Building Video Generation from Floor Plans
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in effectively capturing, representing, and utilizing building interior information, including identifying buildings that meet specific criteria and displaying visual information to remote users, due to limitations in floor plans, textual descriptions, and the difficulty in constructing and maintaining accurate and complete building information.
Innovation Solution
The development of a system that automatically generates building videos based on the analysis of building images and floor plans, using machine learning models to identify building attributes and generate textual descriptions, and allowing for the selection and manipulation of video segments to create new videos or overlay additional data on existing ones.
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, scale, and visualize
Solution Approach 1:
The patent uses computer-generated images and videos as copies of building interiors, created from floor plan data and camera trajectory information. These visual copies replace the need for manual floor plan construction and maintenance while preserving all interior layout information, eliminating the complexity of creating and updating traditional floor plans.
Solution Approach 2:
The patent replaces the mechanical process of manually constructing and maintaining floor plans with an automated computer-based system that generates visual representations algorithmically from structured data and camera trajectories, eliminating manual intervention and associated complexity.
2Loss of information
If textual descriptions of buildings are used, then some building information is available, but they are often inaccurate and incomplete
Solution Approach 1:
The patent creates visual copies of building interiors through computer-generated images and videos that directly represent the actual spatial layout and features. These visual representations provide complete and accurate building information without the inaccuracies and omissions inherent in textual descriptions.
3Loss of information
If building images and floor plans are analyzed manually, then building attributes can be identified, but it is time-consuming and inefficient
Solution Approach 1:
The patent implements automated systems where computer algorithms independently analyze floor plan images, extract building attributes, and generate visual representations without human intervention. The system serves itself by automatically processing data, identifying features, and producing results, eliminating time-consuming manual analysis.
Solution Approach 2:
The patent replaces manual building analysis with automated computer-based image processing and pattern recognition algorithms that rapidly extract building attributes from floor plan data, substituting human time and effort with efficient computational processes.
4Loss of information
If traditional building representation methods are used, then building information can be stored, but it is difficult to display visual information to remote users in user-selected manners
Solution Approach 1:
The patent creates dynamic, interactive video representations of building interiors that can be controlled and manipulated in real-time by remote users. The system allows users to navigate, zoom, and explore different areas of the building virtually, providing flexible and adaptive visual information display that responds to user selections and actions.
Solution Approach 2:
The patent transitions from static floor plan representations to dynamic video and image sequences that add temporal and interactive dimensions. This allows remote users to explore building interiors from multiple perspectives and at different scales, providing comprehensive visual information control that was impossible with traditional 2D representations.
Data Source
Figure 1A
Figure 1B
Figure 2A
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.