Distractor Map Object Tracking for Complex Scenes

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing object tracking systems face challenges in accurately distinguishing between target objects and distractors, especially in complex scenes with high similarity or occlusion, leading to inefficiencies and increased computational costs.

Innovation Solution

The use of a distractor map to represent distractor information in a low-capacity format, allowing for efficient distinction between target objects and distractors through simple operations, reducing computational overhead by using a partial region for tracking and incorporating a neural network-based motion estimation model to update distractor maps.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If existing object tracking systems use traditional methods to distinguish target objects from distractors, then tracking can be performed, but accuracy deteriorates in complex scenes with high similarity or occlusion

Engineering Contradiction:
Improvetracking accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the tracking problem into two distinct components: a target object detector that identifies potential target regions, and a distractor map that represents distractor information in a compact format. This segmentation allows the system to handle complex scenes by separating target detection from distractor identification, improving accuracy without requiring a monolithic complex system.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent changes the representation parameter of distractor information from detailed image data to a low-capacity distractor map that encodes distractor locations and characteristics. This parameter change reduces the computational burden while maintaining the ability to distinguish distractors from targets, resolving the contradiction between accuracy and system complexity.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If existing object tracking systems process full-resolution images to distinguish targets from distractors, then tracking accuracy can be maintained, but computational cost increases

Engineering Contradiction:
Improvetracking accuracyVSAvoidcomputational cost
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent creates a simplified copy of distractor information in the form of a distractor map, which stores essential distractor characteristics in a compressed format. This copy allows the system to reference distractor locations and features without processing full-resolution images, reducing computational cost while maintaining tracking accuracy through the use of this compact representation.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent applies partial action by using only the necessary portion of image data for tracking. Instead of processing complete high-resolution images, the system processes a partial representation through the distractor map that contains only the essential information needed to distinguish distractors from targets, significantly reducing computational requirements.

Inventive Principle:
Principle #16Partial or excessive action

3Measurement precision

If existing object tracking systems use detailed representations of distractor information, then distinction accuracy improves, but computational overhead increases

Engineering Contradiction:
Improvedistractor distinction accuracyVSAvoidprocessing efficiency
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent transforms distractor information from detailed image representations to a low-capacity parameterized format in the distractor map. This parameter change retains the essential characteristics needed for accurate distractor distinction (locations, key features) while dramatically reducing the data volume, thereby improving processing efficiency without sacrificing distinction accuracy.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent segments distractor information into discrete, manageable units within the distractor map structure. By organizing distractor data into a segmented, structured format rather than processing continuous detailed images, the system achieves efficient processing while maintaining the ability to accurately distinguish distractors from target objects.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4202832B1Method and apparatus with object tracking
Publication Date: 2025.10.22 SAMSUNG ELECTRONICS CO LTD
  • EP4202832B1 patent drawingFigure 1
  • EP4202832B1 patent drawingFigure 2
  • EP4202832B1 patent drawingFigure 3

AI summary

A method and apparatus for object tracking are provided, where the object tracking method includes determining box information of candidate boxes in a current image frame and similarity scores of the candidate boxes based on including a search region of the current image frame with a template image corresponding to a target object, adjusting the similarity scores of the candidate boxes using a distractor map including distractor information of a previous image frame, determining a target box corresponding to the target object and a distractor box corresponding to a distractor of the target object from the candidate boxes based on the adjusted similarity scores, and updating the distractor map based on distractor information of the current image frame according to the distractor box.