Skin Reflectance Database for Scene Illuminant Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing augmented reality applications face challenges in accurately rendering virtual objects due to inconsistencies in scene illumination, particularly when juxtaposed with real-world objects, leading to unrealistic or inaccurate color representation.
Innovation Solution
A method and apparatus that estimate scene illuminant based on skin reflectance, using a camera to capture an image, identify human skin regions, determine a representative skin color value, select a candidate skin reflectance spectrum, and update the estimated illuminant spectral power distribution to enhance color accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional illuminant estimation algorithms are used, then the processing speed is fast, but the color accuracy and realism of virtual objects deteriorate
Solution Approach 1:
The patent pre-establishes a skin reflectance database containing spectral reflectance values for different skin types before actual illuminant estimation. This preliminary preparation allows the system to quickly match observed skin colors against pre-computed reference data, achieving high color accuracy without complex real-time spectral analysis.
Solution Approach 2:
The patent creates a simplified spectral model by copying and storing characteristic skin reflectance patterns in a database. Instead of performing complex spectral decomposition during runtime, the system copies pre-analyzed skin reflectance data and matches it against observed skin colors, significantly reducing processing complexity while maintaining accuracy.
2Measurement precision
If spectral reflectance database is used, then the color consistency improves, but the computational time increases
Solution Approach 1:
The patent segments the skin reflectance database into distinct categories (e.g., different skin types, age groups, or reflectance characteristics). This segmentation allows the system to quickly narrow down the search space by first identifying the relevant skin type category, then matching within that subset, thereby reducing computational time while maintaining color consistency.
Solution Approach 2:
The patent transforms the spectral reflectance data into a reduced parameter representation (e.g., using principal component analysis or other dimensionality reduction techniques). By changing the parameter space from full spectral data to a smaller set of dominant parameters, the system achieves fast comparison and matching while preserving the essential color information needed for consistent illuminant estimation.
3Measurement precision
If iterative refinement is performed, then the illuminant estimation accuracy improves, but the processing duration increases
Solution Approach 1:
The patent implements a feedback mechanism where the initial illuminant estimation is used to compute expected skin colors, which are then compared against observed skin colors. The error from this comparison feeds back into refining the illuminant estimate. This controlled feedback loop achieves high accuracy while limiting processing duration by using the feedback iteratively rather than exhaustively.
Solution Approach 2:
The patent performs a limited number of refinement iterations (partial action) rather than continuing until complete convergence. By performing just enough iterations to achieve satisfactory accuracy (e.g., 1-3 refinement steps), the system balances illuminant estimation accuracy with processing efficiency, avoiding the excessive computational cost of exhaustive refinement.
Data Source
AI summary
Some embodiments of a method may include obtaining an image a real-world environment, determining an estimated illuminant spectral power distribution of an illuminant of the real-world environment, and detecting a region of the image representing human skin. The method may further include determining a representative skin color value of the region and based on the estimated illuminant spectral power distribution and the representative skin color value most closely matching the representative color value, selecting a candidate skin reflectance spectrum. The method may further include updating the estimated illuminant spectral power distribution base on the representative skin color value and the selected candidate skin reflectance spectrum.