Automatic Key-Image Database Management for Camera Pose Tracking

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for managing keyframe databases in augmented reality laparoscopic procedures are cumbersome, requiring frequent user intervention and consuming significant hardware resources, leading to suboptimal performance and user experience.

Innovation Solution

A method and device for automated management of keyframes using multi-criteria selection based on reference point matching, pose estimation quality, and image sharpness, allowing new keyframes to be added to the database without impacting real-time video processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual addition of key images is used to complete the 3D model, then tracking coverage is improved, but user independence deteriorates and operation complexity increases

Engineering Contradiction:
Improvetracking coverageVSAvoiduser independence
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The system automatically manages the keyframe database by detecting when current images should be added as new keyframes based on similarity thresholds and tracking performance metrics, eliminating the need for manual user intervention while maintaining comprehensive tracking coverage

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system continuously monitors tracking performance and image similarity metrics, using this feedback to automatically determine when to add new keyframes to the database, creating a closed-loop system that adapts to tracking needs without user input

Inventive Principle:
Principle #23Feedback

2Ease of operation

If automatic selection of new key images is implemented, then user independence is improved, but hardware resource consumption increases

Engineering Contradiction:
Improveuser independenceVSAvoidhardware resource consumption
Core Design Contradiction:
Ease of operationVSUse of energy by moving object

Solution Approach 1:

The system uses adjustable similarity thresholds and selection criteria parameters to control the frequency and conditions under which new keyframes are added, allowing optimization of the balance between automation level and computational resource consumption

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The system implements selective automatic keyframe addition only when specific conditions are met (tracking loss detected, similarity threshold exceeded), rather than continuously processing all images, reducing overall hardware resource consumption while maintaining effectiveness

Inventive Principle:
Principle #16Partial or excessive action

3Measurement precision

If frequent keyframe database updates are performed, then tracking accuracy in changing areas is improved, but processing time increases

Engineering Contradiction:
Improvetracking accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs keyframe database updates periodically based on detected tracking performance degradation or image similarity thresholds rather than continuously, reducing processing overhead while maintaining accuracy in dynamically changing surgical fields

Inventive Principle:
Principle #19Periodic action

Solution Approach 2:

The system proactively adds keyframes before complete tracking loss occurs by monitoring similarity metrics and adding keyframes when thresholds are approached, preventing tracking failures and reducing the need for corrective processing

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP4449356B1Method and device for real-time tracking of the pose of a camera by automatic management of a key-image database
Publication Date: 2025.10.29 SURGAR
  • EP4449356B1 patent drawingFigure 1
  • EP4449356B1 patent drawingFigure 2

AI summary

The invention relates to a method for real-time tracking of the pose of a camera with respect to a 3D model comprising a key-image database that comprises a first execution thread for estimating the current pose of the camera from a selected key image, characterised in that the method further comprises a step of analysing the quality of the estimation of the pose of the camera from the image stored in a buffer memory, and a step of analysing the sharpness of the image stored in the buffer memory from data representative of the movement of the camera, and if the quality of the estimation of the pose of the camera from the stored image and the sharpness of the stored image comply with predetermined criteria, a step of adding the image stored in the buffer memory to the key-image database, as a new key image.