Absolute Scale Depth Calculation Using Geometric Consistency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in deriving absolute scale depth information from captured images, especially in environments where the camera installation height is unknown or fluctuates, and when the methods for acquiring position and posture differ from those for acquiring depth.
Innovation Solution
The absolute scale depth calculation device includes a captured image acquisition unit, an ambiguous scale depth deriving unit, a position and posture acquisition unit, and an absolute scale depth deriving unit. It uses geometrical consistency from multiple captured images to derive absolute scale depth information, even in environments with unknown or fluctuating camera heights.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If camera installation height is measured beforehand to calculate absolute scale depth, then absolute scale depth information can be derived, but the method becomes inapplicable when camera height is unknown or fluctuates
Solution Approach 1:
The system performs self-calibration by automatically determining the camera installation height through image processing and geometric calculations without requiring external measurement tools or manual input. The camera system itself generates the calibration data needed for absolute scale depth calculation.
Solution Approach 2:
The system transitions from requiring a fixed, pre-measured camera height parameter to dynamically calculating the height parameter through image analysis. This allows the system to adapt to fluctuating heights by continuously updating the height parameter based on current imaging conditions.
2Measurement precision
If visual SLAM and GNSS are combined to convert position and posture to absolute scale, then absolute scale position can be obtained, but the method requires identical scales between position/posture and depth information which limits applicability
Solution Approach 1:
The system introduces an intermediary calibration process that mediates between different measurement systems. By using image processing as an intermediate step to establish geometric relationships, the system bridges the scale differences between position/posture acquisition methods and depth information acquisition methods.
Solution Approach 2:
The system separates the depth calculation process into independent stages: first acquiring ambiguous scale depth information from images, then independently determining camera parameters, and finally combining these to produce absolute scale depth. This segmentation allows each component to work with its own scale system.
3Device complexity
If monocular camera is used to capture images for depth calculation, then device complexity is reduced, but the ability to derive absolute scale depth information is compromised
Solution Approach 1:
The system replaces complex multi-camera mechanical setups with a simpler monocular camera system. By substituting the mechanical complexity of multiple sensors with computational algorithms that process single-camera images, the system maintains absolute scale depth capability while reducing device complexity.
Data Source
AI summary
According to an embodiment, an absolute scale depth calculation device includes a captured image acquisition unit, an ambiguous scale depth deriving unit, a position and posture acquisition unit, and an absolute scale depth deriving unit. The captured image acquisition unit is configured to acquire a plurality of captured images at different imaging time points from an imaging unit mounted on a moving body. The ambiguous scale depth deriving unit is configured to derive ambiguous scale depth information from a captured image. The position and posture acquisition unit is configured to acquire absolute scale position and posture information pertaining to the imaging unit when each of the plurality of captured images is captured. The absolute scale depth deriving unit configured to derive absolute scale depth information, based on a geometrical consistency from the plurality of captured images, the ambiguous scale depth information, and the absolute scale position and posture information.


