Subject Identification via Movement Timing Across Multiple Cameras
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
When multiple information processing apparatuses with cameras photograph a subject from different viewpoints, it becomes difficult to extract the same subject due to variations in the orientation of the subject's face in the images.
Innovation Solution
An information processing apparatus and method that includes an image pickup unit, a movement detecting unit, a data generating unit, and a determining unit to detect changes in subject movement, generate timing data, and compare it with data from other apparatuses to determine if the subjects are the same or similar, allowing for accurate identification despite different viewpoints.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple information processing apparatuses photograph a subject from different viewpoints, then more comprehensive subject coverage is achieved, but subject identification accuracy deteriorates due to varying face orientations
Solution Approach 1:
The patent replaces the mechanical/visual system of face recognition with a temporal measurement system. Instead of analyzing facial features and orientations across different viewpoints, the system measures the timing of mouth opening/closing actions. This temporal measurement approach is viewpoint-independent, allowing accurate subject identification even when apparatuses photograph from different angles, thus resolving the contradiction between comprehensive coverage and identification accuracy.
2Reliability
If traditional face recognition methods are used across multiple apparatuses, then subject identification is attempted, but processing complexity increases and real-time identification becomes difficult
Solution Approach 1:
The patent extracts only the essential temporal information (mouth opening/closing timing) from the complex face recognition process. By isolating this specific temporal characteristic and using it as the primary identification criterion, the system dramatically reduces processing complexity while maintaining reliable subject identification across multiple apparatuses.
Solution Approach 2:
The patent changes the identification parameter from spatial features (facial orientation, features) to temporal parameters (timing of mouth movements). This parameter transformation simplifies the identification process and enables real-time operation, as temporal measurements are faster and less computationally intensive than traditional face recognition algorithms.
3Measurement precision
If face images are pre-stored for recognition, then identification accuracy improves, but system complexity and storage requirements increase
Solution Approach 1:
The patent uses temporary temporal measurements (mouth movement timing) instead of permanent stored face images. These temporal data are generated on-demand during photographing and discarded after identification, eliminating the need for extensive databases of pre-stored facial images while maintaining identification accuracy.
Data Source
AI summary
There is provided an information processing apparatus including: an image pickup unit photographing a subject; a movement detecting unit detecting changes in movement of a subject photographed by the image pickup unit; a data generating unit generating first timing data relating to change timing at which the movement detecting unit detected a change in the movement; a receiving unit receiving second timing data, which relates to change timing when movement of a subject photographed by another apparatus was detected, from the other apparatus during photographing by the image pickup unit; and a determining unit determining, based on the first timing data generated by the data generating unit and the second timing data received by the receiving unit, whether the subject photographed by the image pickup unit and the subject photographed by the other apparatus are the same or a similar subject.