Structured Light Depth Map Outlier Correction via Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Structured light systems face challenges in accurately generating depth maps due to reflective or transparent surfaces, dynamic range exceedance, and sub-surface scattering, leading to inconsistencies and errors in depth map detection.
Innovation Solution
An apparatus and method that project and receive codewords, divide the depth map into segments, determine lower and upper quantile depth values, generate a global depth range, identify outlier depth values, and correct them based on neighboring values within the global range, using histogram or gradient calculations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If structured light systems project and receive codewords to generate depth maps, then depth map generation capability is achieved, but outlier depth values appear due to reflections, transparent surfaces, and sub-surface scattering
Solution Approach 1:
The depth map is divided into multiple segments, and quantile depth values are calculated for each segment independently. This segmentation allows local statistical analysis to identify outliers without being affected by global depth variations, thereby improving both measurement precision and reliability.
Solution Approach 2:
The system calculates quantile depth values from the depth map segments and uses these statistical measures as feedback to identify and correct outlier depth values. This feedback mechanism continuously refines the depth map accuracy by comparing actual depth values against statistically derived expected ranges.
2Device complexity
If the system processes the entire depth map globally, then processing is simplified, but local depth variations and outliers cannot be effectively distinguished
Solution Approach 1:
Instead of processing the entire depth map as a single unit, the system segments the depth map into multiple regions and performs quantile calculations on each segment. This approach balances processing complexity by using localized statistics rather than global analysis, while significantly improving outlier detection accuracy through segment-specific depth range determination.
3Measurement precision
If quantile calculations are performed on each segment, then outlier detection accuracy is improved, but processing time and computational load increase
Solution Approach 1:
The depth map is divided into segments that can be processed in parallel, reducing the overall processing time while maintaining high outlier detection accuracy. Each segment's quantile calculations are independent, allowing for efficient computation without the need to analyze the entire depth map sequentially.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach effectively corrects outlier depth values, improving the accuracy and reliability of depth map generation by identifying and adjusting invalid or accidentally valid depth values, thereby enhancing the utility of the depth map.
Implementation Method 1
projecting the plurality of codewords includes propagating light through a codemask
Implementation Method 2
receive a reflection of the projected codewords from a surface
Data Source
AI summary
Systems and methods for correcting errors in a depth map generated by a structured light system are disclosed. In one aspect, a method includes dividing a depth map into segments and calculating a density distribution of the depth values for each segment. The method includes detecting error (or “outlier”) values by determining the depth values that fall outside of a range of depth values, the range of depth values representative of the highest density depth values for a given segment. The method includes detecting error values in the depth map as a whole based on the density distribution values for each segment.