Infinite Repeat Pattern Editing for Distortion-Free AI Modification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image editing tools, including AI models, struggle to adequately modify complex infinite repeat patterns in textile designs, especially those with aspect ratios different from 1:1, leading to time-consuming and expensive processes with undesirable distortions.
Innovation Solution
A method involving the use of AI models to modify a selection area within a modification area, comprising borders, joints, and perimeter points, with tapered sections to align with these features, allowing simultaneous modification along both vertical and horizontal axes without distortion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If existing image editing tools including AI models are used to modify complex infinite repeat patterns, then the modification process becomes time-consuming and expensive, but the patterns still suffer from undesirable distortions and cannot adequately accommodate complex aspect ratios
Solution Approach 1:
The patent segments the infinite repeat pattern into multiple tile units that can be independently processed. Each tile is modified separately using AI models, then the modified tiles are assembled into the complete pattern. This segmentation allows parallel processing of multiple tiles, significantly reducing modification time while maintaining high precision through controlled adjustments at tile boundaries.
Solution Approach 2:
The patent performs preliminary analysis to identify border regions, joints, and perimeter points in the pattern before modification. By pre-mapping these critical areas, the AI model can focus computational resources on precisely modifying only the necessary regions, avoiding unnecessary processing time while ensuring accurate modification results without distortions.
2Adaptability or versatility
If existing AI models are used to modify infinite repeat patterns with aspect ratios different from 1:1, then the designs are limited to simple aspect ratios, but the modification process introduces undesirable distortions
Solution Approach 1:
The patent explicitly handles asymmetric aspect ratios by allowing tile units to have different dimensions rather than forcing a 1:1 square format. The system identifies and modifies tiles according to their actual dimensional relationships, preserving the intended aspect ratio while avoiding distortions through asymmetric tile processing and boundary alignment strategies.
Solution Approach 2:
The patent changes the parameter of tile boundary handling by specifically identifying and processing border regions, joints, and perimeter points differently from interior regions. This parameter-based differentiation allows the AI model to adapt to various aspect ratios while maintaining modification accuracy, as the system adjusts its processing based on the geometric parameters of each tile configuration.
3Adaptability or versatility
If modern software applications are used to create and modify infinite repeat patterns by selecting portions of source images, then design variety and complexity are increased, but the modification process becomes time-intensive and expensive
Solution Approach 1:
The patent creates multiple tile copies from a single modified tile unit. Once one tile is successfully modified with the desired design variety and complexity, the same modification can be rapidly copied and applied to all other tiles in the pattern. This copying approach maintains high design versatility while dramatically improving modification efficiency, as the time-intensive AI processing is performed only once per unique tile design rather than individually for each tile.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Systems and methods for modifying an infinite repeat pattern of a digital image are disclosed. An infinite repeat pattern of a digital source image is obtained, and a modification area is selected that includes a plurality of duplicates of portions of the digital source image. Within the modification area, the borders between the duplicates, the joints between multiple borders, and the points where the borders meet the perimeter of the modification area are determined. The joints are centered along a central axis of the modification area. An additional selection area within the modification area is made based on the earlier-determined borders, joints, and points on the perimeter of the modification area. The additional selection area is modified using an artificial intelligence (Al) model to create an infinite repeat pattern without clearly visible borders and/or boundaries between the duplicates of the digital source image, and without other undesirable features.