Vehicle Camera Orientation Estimation Using Roadside Landmarks

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing camera orientation estimation methods on autonomous vehicles are unreliable due to vibrations and environmental forces, affecting the accuracy of object location determination by cameras.

Innovation Solution

Estimate camera orientation using landmarks detected from images, utilizing a landmark pixel location module to determine the landmark's pixel location and orientation, using a combination of intrinsic and geometric methods, and a landmark pixel location module to calculate the camera's orientation, and a landmark pixel location, and a camera orientation, based on a landmark detected from an image obtained by the camera, and using a landmark pixel location module to determine the landmark's pixel location, and a camera orientation module to calculate the camera's orientation, using an intrinsic matrix and a previously known extrinsic matrix.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional camera orientation estimation methods are used on autonomous vehicles, then the system structure remains simple, but the measurement precision deteriorates due to vibrations and environmental forces affecting camera reliability

Engineering Contradiction:
Improvecamera orientation estimation accuracyVSAvoidcamera orientation reliability under vibration
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent introduces an intermediary computational process that uses the detected landmark's known world coordinates and pixel coordinates to calculate camera orientation parameters. This intermediary calculation method (using intrinsic matrix and geometric relationships) serves as a mediator between the raw image data and the final orientation estimation, enabling accurate orientation determination even when the camera is subjected to vibrations and environmental forces.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces traditional mechanical orientation sensing methods with a computational geometry-based approach. Instead of relying on physical sensors that may be affected by vibrations, the system uses mathematical relationships between landmark coordinates, pixel positions, and camera parameters to compute orientation, substituting mechanical measurement with optical-computational measurement.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If landmark detection methods are implemented to improve orientation estimation accuracy, then measurement precision improves, but device complexity increases due to additional processing modules

Engineering Contradiction:
Improveorientation estimation precisionVSAvoidprocessing system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The landmark pixel location module serves multiple functions: it detects landmarks in images, determines their pixel coordinates, identifies their world coordinates from map data, and uses this information to calculate camera orientation. This multi-functional module reduces the need for separate dedicated components for each function, thereby managing device complexity while achieving high measurement precision.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system uses freely available map data containing landmark world coordinates as a self-service resource. Instead of requiring additional expensive positioning infrastructure, the system leverages existing public data (map data with landmark locations) to enable accurate orientation estimation, reducing the need for additional hardware components.

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP3813020B1Camera orientation estimation
Publication Date: 2025.12.31 CREATEAI INC
  • EP3813020B1 patent drawingFigure 1
  • EP3813020B1 patent drawingFigure 2
  • EP3813020B1 patent drawingFigure 3

AI summary

Techniques are described to estimate orientation of one or more cameras located on a vehicle. The orientation estimation technique can include obtaining an image from a camera located on a vehicle while the vehicle is being driven on a road, determining, from a terrain map, a location of a landmark located at a distance from a location of the vehicle on the road, determining, in the image, pixel locations of the landmark, selecting one pixel location from the determined pixel locations; and calculating values that describe an orientation of the camera using at least an intrinsic matrix and a previously known extrinsic matrix of the camera, where the intrinsic matrix is characterized based on at least the one pixel location and the location of the landmark.