Abstract Texture Generation for Building Facades
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Providers of mapping-related services face challenges in efficiently creating textures that realistically represent real-world objects like buildings, as traditional methods are resource-intensive and struggle with vast numbers of objects, especially in three-dimensional mapping, where manual effort and raw texture images suffer from quality issues like shadows and occlusions.
Innovation Solution
A computer-implemented method and apparatus for generating abstract texture data by receiving pixel-level labeling data, processing it to determine window size and spacing, and computing a confidence score to create a window pattern, which is then used to generate an abstract texture for building facades, addressing the complexity of varying structures and image defects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual methods are used to create textures for building models, then the visual quality and realism of the texture can be maintained, but the resource burden and time consumption increase significantly
Solution Approach 1:
The patent uses raw texture images captured from the real world as templates to generate abstract textures for building models. By copying visual characteristics from real building images and applying them to 3D models, the system maintains visual quality while eliminating manual texture creation efforts for each building
Solution Approach 2:
The system automatically processes raw texture images through image processing algorithms to generate abstract textures without requiring manual intervention. The computational system performs the texture generation task itself, reducing the resource burden of manual texture creation while maintaining visual quality
2Reliability
If raw texture images are used directly for building models, then the visual quality can be maintained, but quality issues like shadows and occlusions persist
Solution Approach 1:
The patent extracts only the essential visual characteristics from raw texture images, such as window patterns, wall colors, and architectural features, while leaving out problematic elements like shadows, occlusions, and lighting artifacts. This extraction process maintains the representative visual quality while eliminating harmful factors
Solution Approach 2:
The system uses image processing algorithms to identify and correct distortions in raw texture images, converting the harmful effects of shadows and occlusions into opportunities to generate cleaner, more accurate abstract textures that represent the true building characteristics
3Reliability
If detailed textures are created for every building, then the visual realism is improved, but the data storage requirements increase significantly
Solution Approach 1:
The patent applies abstract textures that capture the essential visual characteristics of building facades without storing complete high-resolution images. By focusing on key features like window patterns and wall colors at the appropriate level of detail, the system maintains visual realism while significantly reducing data storage requirements compared to storing full raw texture images for every building
Data Source
AI summary
An approach involves receiving pixel-level labeling data for an image depicting at least a portion of a building facade. The pixel-level labeling data labels each of a plurality of pixels of the image as either window pixels or non-window pixels. The approach further involves generating a window pattern based on window size data, window spacing data, or a combination thereof extracted from the pixel-level labeling data. The approach further involves computing a confidence score for the window pattern based on at least one observed value of at least one characteristic of the window pattern or a deviation of the observed value from at least one expected value, and generating an abstract texture based on the window pattern, the confidence score, and/or other probabilistic metrics.