Reference-Guided Image Expansion for Accurate Scene Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image expansion methods using generative artificial intelligence often generate expanded content that does not conform to common sense, leading to low accuracy.
Innovation Solution
An image processing method that utilizes a target model to generate a display image by aligning and expanding a target image based on a reference image captured at the same location and orientation, ensuring the same perspective and scene, thereby improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If generative artificial intelligence is used for image expansion, then image expansion capability is achieved, but the accuracy and common sense conformity of expanded content deteriorates
Solution Approach 1:
The patent applies preliminary action by capturing a reference image with a wider field of view before performing image expansion. This reference image is processed in advance to extract feature information, which is then used to guide the expansion process. The target image is expanded based on this pre-prepared reference data, ensuring that the expanded content aligns with real-world common sense and maintains high accuracy.
2Measurement precision
If a target model is used to generate display image from reference image and target image, then the accuracy and completeness of expanded content is improved, but the device complexity increases
Solution Approach 1:
The patent introduces a target model as an intermediary component that processes both the reference image and target image. This target model acts as a mediator that integrates information from both images, extracts relevant feature information, and generates the expanded display image. By using this intermediary processing mechanism, the system achieves high accuracy in expanded content while managing complexity through a dedicated processing module.
Data Source
AI summary
An image processing method includes: obtaining a target image and a reference image, the target image having a smaller field of view than the reference image; and using a target model to generate a display image from the reference image and the target image, where the display image includes image content of the reference image and is larger than the target image.