Texture Extraction from Text-Based Images Using Kerning Adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional digital graphics systems cannot extract textures from text in digital images, limiting the ability to replicate or modify text designs effectively.
Innovation Solution
A system that identifies and segments glyphs from digital images using text recognition and segmentation, adjusts kerning values to reduce gaps between glyphs, and generates a synthesized texture through image inpainting, allowing the extraction and application of textures to target digital text objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional text recognition systems are used, then text can be identified, but texture extraction from text is not possible
Solution Approach 1:
The patent extracts texture information from text images by separating the texture component from the text structure. The system identifies text regions, extracts texture samples from these regions, and creates standalone texture representations that can be applied to other text objects, effectively taking out the texture property from the original text image.
Solution Approach 2:
The patent segments the text image into individual text elements or regions to analyze and extract texture characteristics from each segment. By dividing the text image into manageable portions, the system can effectively capture texture variations across different text elements and synthesize a comprehensive texture representation.
2Measurement precision
If deep learning models are used for texture extraction, then accurate texture representation is achieved, but computational resources and cloud service dependency increase
Solution Approach 1:
The patent creates simplified copies or representations of texture information from text images. Instead of using complex deep learning models, the system generates texture samples that capture the essential visual characteristics and applies these samples to target text objects, achieving accurate texture representation with reduced computational overhead.
Solution Approach 2:
The patent transforms the texture extraction problem by changing the approach parameters - moving from model-based deep learning to sample-based extraction. The system adjusts extraction parameters such as sample size, resolution, and application methods to achieve accurate texture representation while minimizing computational resource consumption.
Data Source
AI summary
This disclosure describes one or more implementations of systems, non-transitory computer-readable media, and methods that extract a texture from embedded text within a digital image utilizing kerning-adjusted glyphs. For example, the disclosed systems utilize text recognition and text segmentation to identify and segment glyphs from embedded text depicted in a digital image. Subsequently, in some implementations, the disclosed systems determine optimistic kerning values between consecutive glyphs and utilize the kerning values to reduce gaps between the consecutive glyphs. Furthermore, in one or more implementations, the disclosed systems generate a synthesized texture utilizing the kerning-value-adjusted glyphs by utilizing image inpainting on the textures corresponding to the kerning-value-adjusted glyphs. Moreover, in certain instances, the disclosed systems apply a target texture to a target digital text based on the generated synthesized texture.