Camera Parameter Identification Using Skeleton-Based Height Cues
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems struggle to accurately estimate camera parameters when the height of a person in an image is unknown, leading to challenges in calculating feature amounts such as height from video data.
Innovation Solution
A non-transitory computer-readable recording medium stores an identification program that acquires image data, generates skeleton and attribute information, sets feature amounts based on these data, and optimizes camera parameters to minimize the difference between estimated and actual feature amounts, enabling accurate calculation of person heights even when the person's height is unknown.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If camera parameters are estimated without known person height, then the system can process images of unknown persons, but the estimation accuracy of camera parameters deteriorates
Solution Approach 1:
The system uses feedback from skeleton-based height estimation to iteratively refine camera parameter estimation. The skeleton information provides a feedback loop that allows the system to adjust and optimize camera parameters without requiring prior knowledge of person height, thereby resolving the contradiction between processing unknown persons and maintaining estimation accuracy.
Solution Approach 2:
Skeleton information acts as an intermediary between the image data and the camera parameter estimation. By introducing skeleton-based attribute information as a mediator, the system can bridge the gap between unknown person dimensions and camera parameters, enabling accurate estimation without direct measurement of person height.
2Productivity
If skeleton information is used to estimate person height, then the system can calculate feature amounts, but the accuracy deteriorates when camera parameters are unknown
Solution Approach 1:
The system establishes a feedback mechanism where skeleton-based height estimates are continuously refined using camera parameter optimization. This iterative feedback process allows the system to maintain high accuracy in feature amount calculation while being able to process images of unknown persons, resolving the contradiction between productivity and precision.
Solution Approach 2:
The system dynamically adjusts camera parameters based on the relationship between skeleton information and image measurements. By changing and optimizing camera parameters iteratively, the system can maintain accurate feature amount calculation even when initial camera parameters are unknown, thus resolving the accuracy deterioration problem.
Data Source
AI summary
A non-transitory computer-readable recording medium stores an identification program for causing a computer to execute a process including: acquiring image data captured by a camera; generating skeleton information of a person included in an image frame and attribute information of the person included in the image frame by analyzing the image frame that constitutes the acquired image data; setting a first feature amount related to a height of the person for the person in the image frame based on the generated attribute information of the person; setting a second feature amount related to the height of the person for the person in the image frame based on the generated skeleton information; and specifying a parameter of the camera based on a difference between the first feature amount and the second feature amount.