Multi-Layer Homography Mapping for Occluded Foot Position Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The accuracy of a person's position on a map decreases when their feet are hidden, leading to errors in coordinate conversion using a homography matrix in existing technologies.
Innovation Solution
Employing multilayer homography matrices, including a homography matrix for the floor surface and additional matrices for different heights, to accurately convert coordinates of visible body parts, and determining whether the feet are hidden, using an information processing apparatus to select the appropriate homography matrix based on the height of the visible body parts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single homography matrix is used to convert coordinates of all body parts, then the conversion process is simple, but the mapping precision deteriorates when feet are hidden
Solution Approach 1:
The patent divides the homography matrix conversion process into multiple layers based on body part height. A first homography matrix is used for lower body parts (feet, legs) and a second homography matrix is used for upper body parts (torso, head), allowing each matrix to be optimized for its specific body region and reducing mapping errors when feet are hidden
2Productivity
If coordinate conversion is performed without detecting hidden body parts, then the processing speed is fast, but the position accuracy deteriorates
Solution Approach 1:
The patent performs preliminary detection of hidden body parts by checking the vertical position of detected body parts relative to each other. Before performing coordinate conversion, the system determines whether the feet are hidden based on the vertical distance between the lowest detected body part and the expected foot position, allowing for accurate position mapping even when feet are not directly visible
Data Source
AI summary
A non-transitory computer-readable recording medium stores therein an analysis program that causes a computer to execute a process including specifying a plurality of portions of a person included in an image configuring a video obtained by capturing in a facility by analyzing the video, converting each coordinate position on the image of the plurality of specified portions of the person into each coordinate position on a map in the facility, extracting any target portion among the plurality of portions based on each positional relationship of the converted coordinate positions on the map, and setting the coordinate position of the extracted target portion on the map as a position where the person exists.