Active IR Stereo Depth Mapping Without Interference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional stereo algorithms are limited in generating depth maps due to the quantity of relevant features and brightness constancy assumptions, and structured light technologies face interference and calibration issues when multiple modules sample the same scene simultaneously.
Innovation Solution
The method employs active IR stereo modules that project an infrared dot pattern and use synchronized IR cameras to compute disparity maps, generating depth maps without interference by treating each module's pattern as random, ensuring temporal coherence and accurate depth mapping even in non-studio conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional stereo algorithms are used to generate depth maps from RGB images, then the process is simple and widely applicable, but the effectiveness is limited by the quantity of relevant features and brightness constancy assumptions, making it difficult to generate depth maps for solid color objects or in non-studio conditions
Solution Approach 1:
The patent transitions from using RGB color information to using infrared (IR) color information. By capturing images in the infrared spectrum and detecting IR reflectance patterns, the system can identify surface features and material properties that are not visible in the visible spectrum, enabling accurate depth mapping for solid color objects and in non-studio lighting conditions where traditional RGB stereo algorithms fail
Solution Approach 2:
The patent changes the fundamental parameter used for depth estimation from visible light reflectance to infrared reflectance. This parameter change allows the system to penetrate through materials and detect thermal emission patterns, providing robust depth information regardless of visible light conditions, object color, or lighting environment
2Productivity
If multiple structured light modules are used to sample the same scene simultaneously, then the coverage and resolution of depth mapping is improved, but interference occurs between the projected patterns and calibration becomes difficult
Solution Approach 1:
The patent segments the infrared spectrum into multiple bands and assigns different modules to capture different spectral segments. Each module captures a specific portion of the infrared spectrum, allowing simultaneous operation without interference. The system then combines these segmented spectral data to create a complete depth map, achieving high productivity while eliminating module interference
Solution Approach 2:
The patent introduces spectral decomposition as an intermediary mechanism between multiple modules. By processing the infrared signals through spectral analysis, the system separates and identifies contributions from different modules, enabling simultaneous multi-module operation while maintaining clear, interference-free depth information through mathematical separation of spectral components
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 provides accurate and interference-free depth mapping in scenes with inconsistent lighting and low-feature areas, enabling the use of multiple modules to generate a constructive three-dimensional view without the need for precise calibration.
Implementation Method 1
projecting an infrared (IR) dot pattern onto a scene and photographing the pattern by a single IR camera
Implementation Method 2
capturing stereo images from each of two or more synchronized IR cameras
Data Source
AI summary
Methods and systems for generating a depth map are provided. The method includes projecting an infrared (IR) dot pattern onto a scene. The method also includes capturing stereo images from each of two or more synchronized IR cameras, detecting a number of dots within the stereo images, computing a number of feature descriptors for the dots in the stereo images, and computing a disparity map between the stereo images. The method further includes generating a depth map for the scene using the disparity map.