Object Tracing Through Within-Camera Trajectory Pre-Merging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing person tracking techniques struggle with trajectory discontinuity when individuals disappear from view or move between camera frames, and fail to merge trajectories accurately across multiple cameras.
Innovation Solution
An object tracing apparatus that utilizes a tracking unit to assign track IDs to detected regions and a tracing unit to associate these IDs with object IDs based on coexistence possibilities, ensuring accurate trajectory merging by considering imaging time and feature similarities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If movement trajectories are merged sequentially from multiple cameras, then trajectory continuity is improved, but inconsistency in merging results occurs when multiple trajectories from the same camera exist simultaneously
Solution Approach 1:
The patent performs preliminary merging of movement trajectories within each camera before merging between cameras. This preliminary action organizes the trajectories in a structured manner, grouping trajectories from the same camera together, which facilitates subsequent between-camera merging and resolves inconsistencies by ensuring that trajectories are properly associated before cross-camera integration.
Solution Approach 2:
The patent segments the trajectory merging process into two distinct stages: (1) within-camera merging to associate trajectories from the same camera, and (2) between-camera merging to associate trajectories across different cameras. This segmentation allows each stage to focus on specific merging challenges, improving overall accuracy by addressing inconsistency at the appropriate level.
2Adaptability or versatility
If clustering method is used with restriction that trajectories in the same camera are not determined to be of the same person, then between-camera merging is achieved, but inconsistency in within-camera merging cannot be resolved
Solution Approach 1:
The patent applies preliminary within-camera merging before applying the clustering method for between-camera merging. This preliminary action ensures that trajectories from the same camera are correctly associated, resolving within-camera inconsistencies. The clustering method then operates on this pre-organized data, maintaining multi-camera compatibility without compromising within-camera merging reliability.
3Device complexity
If only similarity between two movement trajectories is considered for resolving inconsistency, then processing is simplified, but inconsistency removal processing for multiple trajectories of one person cannot be executed
Solution Approach 1:
The patent segments the inconsistency resolution process into within-camera and between-camera components. By considering similarity within each segment (camera) separately and then integrating results, the system can handle multiple trajectories of one person accurately without overwhelming complexity. Each segment processes a manageable subset of trajectories, maintaining reasonable complexity while achieving precise inconsistency resolution.
Data Source
AI summary
This invention provides an object tracing apparatus that traces an object in a video obtained by an image capturing apparatus, comprising a tracking unit that tracks the object in a frame in the video and obtains track data indicating consecutive appearances of the object, and a tracing unit that generates trace information indicating which object ID each track ID belongs to, by associating a track ID obtained via tracking by the tracking unit with an object ID for uniquely identifying an object, wherein the tracing unit, on a basis of a coexistence possibility relating to imaging time between a track ID associated with one or more existing object IDs and target track data without an associated object ID, determines which object ID the target track data belongs to.