Flow Vector Plausibility Check via Prediction Hypothesis Comparison

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for assessing flow vectors in optical flow analysis lack robustness and reliability, particularly in distinguishing between actual and incorrect movement hypotheses, leading to potential errors in object tracking and camera movement differentiation.

Innovation Solution

A method and device that determine a prediction vector based on feature movement between images, generate a hypothesis vector for future image movement, calculate the similarity between prediction and hypothesis vectors, and evaluate the plausibility of the hypothesis vector, considering camera movement and object continuity, to assess the correctness of flow vector hypotheses.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If existing methods are used to assess flow vectors in optical flow analysis, then the assessment can be performed with simple algorithms, but the robustness and reliability of the assessment is insufficient

Engineering Contradiction:
Improverobustness and reliability of flow vector assessmentVSAvoidcomplexity of assessment algorithm
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by determining a prediction vector based on feature movement between a first and second image before assessing the hypothesis vector. This prediction vector serves as a reference that is established in advance to evaluate the plausibility of subsequent flow vector hypotheses, improving reliability without requiring complex real-time analysis

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements feedback by using the prediction vector as a reference against which hypothesis vectors are continuously evaluated. The degree of similarity calculation provides feedback on the plausibility of each hypothesis, allowing the system to iteratively refine its assessment and distinguish actual movement from incorrect hypotheses

Inventive Principle:
Principle #23Feedback

2Measurement precision

If simple algorithms are used for flow vector assessment, then the processing speed is fast, but the ability to distinguish between actual and incorrect movement hypotheses is poor

Engineering Contradiction:
Improveprecision in distinguishing actual vs incorrect movement hypothesesVSAvoidprocessing speed of flow vector assessment
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The prediction vector is determined in advance based on observed feature movement, establishing a reference framework before hypothesis evaluation begins. This preliminary setup enables precise distinction between actual and incorrect hypotheses during processing without requiring complex real-time computations for each hypothesis

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent creates a reference copy of the prediction vector that represents the expected movement pattern. This reference is then used to compare against multiple hypothesis vectors, enabling precise differentiation between correct and incorrect hypotheses through systematic comparison rather than complex individual analysis of each hypothesis

Inventive Principle:
Principle #26Copying

Data Source

PatentEP3673459B1Method and device for checking the plausibility of a flow vector hypothesis
Publication Date: 2025.01.15 ROBERT BOSCH GMBH
  • EP3673459B1 patent drawingFigure 1
  • EP3673459B1 patent drawingFigure 2~3
  • EP3673459B1 patent drawingFigure 4~5

AI summary

The invention relates to a device and a method for checking the plausibility of a flow vector hypothesis. This involves the following steps: determining a prediction vector (12), associated with a feature (4), based on a movement of the feature (4) between a first image (1) and a second image (2), wherein the second image (2) is an image following the first image (1) in a sequence of images (20); generating a hypothesis vector (13) associated with the feature (4), which describes a probable movement of the feature (4) between the second image (2) and a third image (3), wherein the third image (3) is an image following the second image (2) in a sequence of images (20); calculating a degree of similarity between the prediction vector (12) and the hypothesis vector (13) based on a difference between the prediction vector (12) and the hypothesis vector (13); and evaluating a plausibility of the hypothesis vector (13) based on the calculated degree of similarity, wherein it is evaluated whether the hypothesis vector (13) describes an actual movement of the feature (4) between the second image (2) and the third image (3).