Face Exposure Correction Using Spatial and Chrominance Weights

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing image processing techniques often result in under-exposed faces due to inadequate exposure correction, especially when lighting is behind the subject, and global stylization filters can degrade skin tones, leading to poor visual results with artifacts like abrupt color transitions.

Innovation Solution

An image processing application that uses spatial and color weights to correct luminance and chrominance in face regions, applying exposure correction based on median luminance and chrominance values to ensure smooth, visually pleasing results, and adjusts stylization to maintain natural skin tones.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If global stylization filters are applied to enhance photographs, then visual enhancement is improved, but skin tones and face exposure are degraded

Engineering Contradiction:

Improvevisual enhancementVSAvoidskin tone accuracy

Core Design Contradiction:

Ease of manufactureVSManufacturing precision

Solution Approach 1:

The patent divides the image processing into two distinct segments: global stylization filtering and localized face region processing. Face regions are identified and segmented from the rest of the image, allowing different processing rules to apply to different regions. This enables global aesthetic enhancement while preserving natural skin tones in face areas through separate, more careful processing.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies different processing qualities to different regions of the image. Global stylization filters are applied to non-face regions for visual enhancement, while face regions receive localized processing that prioritizes natural skin tone preservation. This local quality approach allows the system to optimize for enhancement where appropriate while maintaining accuracy where needed.

Inventive Principle:

Principle #3Local quality

2Measurement precision

If luminance histograms are used to correct exposure, then exposure correction is achieved, but visual smoothness and natural appearance are degraded due to artifacts

Engineering Contradiction:

Improveexposure correctionVSAvoidvisual smoothness

Core Design Contradiction:

Measurement precisionVSManufacturing precision

Solution Approach 1:

The patent introduces weights as an intermediary mechanism between exposure correction and the final image output. These weights are computed based on spatial distance from face centers and chrominance similarity to skin tones, acting as a mediator that smoothly blends corrected and uncorrected regions. This intermediary approach avoids the abrupt transitions and artifacts that would result from direct, unweighted application of histogram-based correction.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent dynamically changes multiple parameters including spatial distance, chrominance similarity, and weight values to achieve smooth exposure correction. By continuously adjusting these parameters across different image regions rather than applying uniform correction, the system maintains visual smoothness while achieving accurate exposure correction in face regions.

Inventive Principle:

Principle #35Parameter changes

3Measurement precision

If skin pixels are isolated and luminance histograms are applied, then exposure correction is attempted, but detection accuracy and result satisfaction are reduced

Engineering Contradiction:

Improveexposure correctionVSAvoidskin region detection

Core Design Contradiction:

Measurement precisionVSDifficulty of detecting and measuring

Solution Approach 1:

The patent performs preliminary face detection and chrominance calculation before applying exposure correction. By first identifying face regions and computing median chrominance values, the system establishes accurate reference points that guide subsequent processing. This preliminary action ensures that exposure correction is applied to the correct regions with proper reference values, improving both detection accuracy and result satisfaction.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent replaces simple skin pixel isolation with a more sophisticated approach using chrominance analysis and spatial weighting. Instead of relying solely on color thresholding to detect skin regions, the system uses chrominance distance calculations and weight-based blending, which are more robust to variations in lighting and skin tones, thereby improving detection accuracy.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS9672414B2Enhancement of skin, including faces, in photographs

Publication Date: 2017.06.06 ADOBE INC

AI summary

An image processing application performs improved face exposure correction on an input image. The image processing application receives an input image having a face and ascertains a median luminance associated with a face region corresponding to the face. The image processing application determines whether the median luminance is less than a threshold luminance. If the median luminance is less than the threshold luminance, the application computes weights based on a spatial distance parameter and a similarity parameter associated with the median chrominance of the face region. The image processing application then computes a corrected luminance using the weights and applies the corrected luminance to the input image. The image processing application can also perform improved face color correction by utilizing stylization-induced shifts in skin tone color to control how aggressively stylization is applied to an image.