Camera Pose Estimation from Overhead and Ground-View Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing advanced driver assistance systems (ADAS) face challenges in accurately determining the pose of a camera mounted on a vehicle relative to an overhead image, which affects the precision of vehicle navigation and control systems.
Innovation Solution
A computer system projects an overhead feature map to multiple candidate poses and calculates feature differences with an observed ground-view map to determine the camera's estimated pose, using machine-learning algorithms to refine the accuracy and actuate vehicle systems like propulsion, braking, and steering.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional pose determination methods are used, then the system complexity is low, but the measurement precision of camera pose is insufficient
Solution Approach 1:
The overhead feature map is segmented into multiple candidate pose regions, each representing a possible camera position and orientation. This segmentation allows the system to evaluate multiple hypotheses simultaneously, improving pose determination accuracy by comparing features across different candidate poses rather than relying on a single traditional method.
Solution Approach 2:
The patent introduces an intermediary feature matching process that compares observed ground-view features with projected features from candidate poses. This intermediary comparison mechanism serves as a bridge between traditional pose methods and the desired high-precision determination, enabling accurate pose estimation without directly implementing complex measurement systems.
2Measurement precision
If multiple candidate poses are evaluated, then the measurement precision improves, but the loss of time increases
Solution Approach 1:
The system performs preliminary actions by pre-generating multiple candidate poses and their corresponding projected feature maps before final pose determination. This preliminary evaluation allows the system to quickly eliminate unlikely candidates and focus computational resources on the most probable poses, reducing overall computation time while maintaining high precision.
Solution Approach 2:
The patent implements a feedback mechanism where the system evaluates feature differences between observed and projected maps for each candidate pose, then uses this feedback to refine and select the optimal pose. This iterative feedback process converges quickly to the correct pose, balancing the need for high accuracy with acceptable computation time by avoiding exhaustive search of all possible poses.
Data Source
AI summary
A computer includes a processor and a memory, and the memory stores instructions executable by the processor to generate an overhead feature map from an overhead image of a geographic area; generate an observed ground-view feature map from a ground-view image captured by a camera within the geographic area, the camera oriented at least partially horizontally while capturing the ground-view image; for each of a plurality of candidate poses of the camera, project the overhead feature map to a ground view defined by the respective candidate pose, resulting in a projected ground-view feature map for each candidate pose; for each projected ground-view feature map, determine a feature difference between the observed ground-view feature map and that projected ground-view feature map; and determine an estimated pose of the camera based on the feature differences.