Tracking Moving Objects via Reference Object Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for monitoring large areas using multiple cameras face difficulties in accurately tracking moving objects when delays occur between cameras or when moving means like elevators or escalators are involved, as they struggle to estimate object appearance based on time and distance.
Innovation Solution
An information processing system that includes input, selection, and estimation means to track moving objects by selecting reference objects with distinct features and estimating the time of object appearance in other cameras based on their movement, even when speeds vary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If time and distance based estimation is used to predict object appearance, then the system can provide advance warning of object appearance, but the accuracy deteriorates when delays occur between cameras or when moving means like elevators or escalators are involved
Solution Approach 1:
The patent introduces reference objects as intermediary elements to mediate between the tracking target and the estimation system. Instead of directly estimating the tracking target's appearance time based on time and distance, the system uses reference objects whose movement patterns are observed to infer the tracking target's behavior. This intermediary approach allows the system to adapt to varying conditions (delays, elevators, escalators) without requiring direct measurement of the tracking target's speed and distance.
Solution Approach 2:
The system dynamically changes the parameters used for estimation based on observed movement patterns. Instead of using fixed time and distance parameters, the system adjusts its estimation based on the actual movement characteristics of reference objects, including variations in speed, delays between camera views, and the presence of moving means like elevators or escalators. This parameter adaptation resolves the contradiction by making the estimation accurate under varying conditions while still providing advance warning.
2Area of stationary object
If the system tracks moving objects using multiple cameras, then it can monitor large areas, but the complexity of object association and tracking increases when speeds vary
Solution Approach 1:
Reference objects serve as intermediaries that simplify the tracking process across multiple cameras. Instead of directly tracking and associating the tracking target across different camera views (which becomes complex when speeds vary), the system uses reference objects to bridge the gap between cameras. By observing reference object movement patterns, the system can infer tracking target behavior without complex direct association, thereby reducing system complexity while maintaining large area coverage.
Solution Approach 2:
The system creates a simplified model or copy of the tracking behavior by observing reference objects. Rather than directly modeling the complex movement of the tracking target across multiple cameras with varying speeds, the system copies the movement patterns of reference objects and uses these copies to predict tracking target behavior. This copying approach reduces the complexity of inter-camera association while maintaining effective tracking across large monitored areas.
Data Source
AI summary
[Problem] To provide an information processing system, an information processing method, and a program capable of suitably tracking a moving object even when there is variation in the speed of the moving object to be tracked.[Solution] Provided are: a video acquisition unit (201) for accepting input of a video taken by a plurality of photographing devices; a reference object selection unit (207) for selecting, from among moving objects seen in the video taken by a first photographing device among the plurality of photographing devices, another moving object different from the moving object to be tracked and seen in the video taken by the first photographing device; and a target object appearance prediction unit (213) for predicting, based on whether the other moving object appeared in the video taken by a second photographing device among the plurality of photographing devices, a time of day at which the moving object to be tracked will appear.