Image Look Transfer via Wavelet Texture Decomposition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for adjusting the contrast and brightness of an image to match a model image are time-consuming and fail to effectively highlight or fade details, making them unsuitable for real-time applications like video feeds and requiring excessive manual effort.
Innovation Solution
The method involves decomposing images using wavelet transforms to analyze and modify texture, allowing for efficient transfer of the look from a model image to a target image by computing texture measures and applying ratios to recompose the image, which can be done in real-time without excessive memory usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If histogram matching is performed on base layer and detail layer using bilateral filter, then the look transfer quality is improved, but the processing time increases significantly
Solution Approach 1:
The patent segments the image processing into distinct frequency components using wavelet transforms, separating images into approximation coefficients (low-frequency) and detail coefficients (high-frequency). This segmentation allows independent processing of different image aspects, enabling efficient look transfer by applying histogram matching selectively to approximation coefficients while preserving detail coefficients, thus reducing overall processing time while maintaining quality.
Solution Approach 2:
The patent changes the processing approach by working in the wavelet transform domain rather than the spatial domain. By transforming images into wavelet coefficients and performing histogram matching in this transformed space, the computational complexity is reduced, enabling real-time processing while maintaining the quality benefits of layer-based processing.
2Manufacturing precision
If manual histogram manipulation is performed to adjust contrast and brightness, then the look matching is improved, but the processing time and manual effort increase
Solution Approach 1:
The patent implements automated look transfer by computing histograms from wavelet transform coefficients and automatically determining the mapping between source and target histograms. The system self-adjusts contrast and brightness parameters through algorithmic histogram matching without requiring manual intervention, thereby achieving accurate look matching while eliminating time-consuming manual manipulation.
Solution Approach 2:
The patent transforms the image data into the wavelet domain where histogram operations become more computationally efficient. By performing histogram calculation and matching on wavelet coefficients rather than raw pixel values, the system reduces the computational burden of manual-like adjustments while maintaining precision in contrast and brightness modification.
3Illumination intensity
If traditional contrast adjustment techniques are applied, then the brightness distribution is improved, but the detail highlighting and fading is insufficient
Solution Approach 1:
The patent segments image processing into frequency-based components through wavelet transforms, separating brightness control (applied to approximation coefficients) from detail preservation (maintained through detail coefficients). This segmentation enables independent optimization of brightness distribution while preserving detail information, allowing accurate highlighting and fading of specific image regions without losing detail fidelity.
Solution Approach 2:
The patent applies local quality enhancement by performing histogram matching on wavelet coefficients that represent different spatial frequencies and locations. This allows different processing strategies to be applied to different regions and frequency bands, enabling precise control over which details are highlighted or faded while maintaining appropriate brightness distribution across the entire image.
Data Source
AI summary
Systems, methods, and apparatus, including computer program products, are provided for changing the look of an image. In some implementations a computer-implemented method is provided. The method includes decomposing a first image using a first plurality of wavelet transforms, each first wavelet transform creating a first transformation resolution, each first transformation resolution having respective first data, and decomposing a second image using a second plurality of wavelet transforms, each second wavelet transform creating a second transformation resolution, each second transformation resolution having respective second data. The method also includes determining a first measure of texture for each first transformation resolution's respective data, applying each first transformation resolution's first measure of texture to each corresponding second transformation resolution's data, and recomposing the second image based on the modified second transformation resolutions.