Object Tracking Correction Through Recently Lost Object Matching

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current object tracking methods in images suffer from anomalies such as multiple tracklets corresponding to the same object, swapped identifiers, and incorrect assignment of identifiers to non-existent objects like shadows or reflections, leading to inefficiencies and errors.

Innovation Solution

A method that assigns a new identifier to newly detected objects and compares them to recently lost objects within a threshold period to confirm their identity, using appearance and position comparisons, with optional operator validation to correct tracking errors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a new identifier is assigned to every newly detected object, then the tracking system can identify and follow multiple objects, but it creates multiple tracklets for the same object when the object is temporarily obscured or leaves the field of view

Engineering Contradiction:
Improvetracking accuracyVSAvoidobject identity continuity
Core Design Contradiction:
ReliabilityVSLoss of information

Solution Approach 1:

The system performs preliminary comparison between newly detected objects and recently lost objects before finalizing identifier assignment. By checking if a new object matches any recently lost object within a threshold period, the system proactively prevents the creation of duplicate tracklets, ensuring object identity continuity while maintaining tracking accuracy.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If the system continuously monitors and compares objects to maintain accurate tracking, then tracking precision improves, but computational complexity and processing time increase

Engineering Contradiction:
Improveobject identification accuracyVSAvoidtracking system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system applies comparison operations selectively rather than universally. It only compares newly detected objects with recently lost objects within a specific time threshold, rather than comparing all objects with all other objects. This localized approach maintains high identification accuracy while significantly reducing computational complexity.

Inventive Principle:
Principle #3Local quality

3Productivity

If the tracking system assigns identifiers based solely on current detection, then processing speed is maintained, but errors occur when objects are temporarily obscured or when shadows/reflections are detected

Engineering Contradiction:
Improvetracking processing speedVSAvoidobject detection accuracy
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system incorporates feedback by comparing newly detected objects against the history of recently lost objects. This feedback mechanism allows the system to verify whether a newly detected object is actually a return of a previously lost object or a genuine new object, thereby reducing detection errors while maintaining processing efficiency through targeted comparisons.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP4592953A1Method and device for correcting object tracking, and method and system for tracking object(s) implementing such correction
Publication Date: 2025.07.30 BULL SA
  • EP4592953A1 patent drawingFigure 1~2
  • EP4592953A1 patent drawingFigure 3~4
  • EP4592953A1 patent drawingFigure 5~7

AI summary

The invention relates to a method (100) for correcting the tracking of object(s) in images, said tracking comprising an allocation of an identifier to each new object used to constitute a tracklet associated with said object and retained as long as said object has not been lost during said tracking; said correction method (100) comprising, for at least one new object detected on a current image, a step (104) of comparing said new object to at least one recently lost object: - which has been tracked, and - whose tracking has been lost for a period less than a predetermined threshold period before the capture of said current image; to confirm or not that it is a new object. It also relates to a computer program and a device implementing such a method, as well as a method and a system for tracking objects implementing such an object tracking correction.