Repositioning Objects Along Trajectories in Image Sequences

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing image editing tools struggle to combine aspects of multiple images from a sequence effectively, particularly in repositioning objects along their movement trajectories, and do not utilize phase-based video motion processing to exaggerate subtle movements.

Innovation Solution

A method implemented by a computing device that receives user input to select an object from a sequence of images, determines its movement trajectory, and repositions the object based on user input, optionally applying a phase-based video motion processing algorithm to exaggerate movements, and combines the repositioned object with a base image to create a new image.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a phase-based video motion processing algorithm is applied to determine exaggerated movements, then subtle movements can be enhanced and made more visible, but the complexity of the processing increases

Engineering Contradiction:
Improvedetection of subtle movementsVSAvoidprocessing algorithm complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces a phase-based video motion processing algorithm as an intermediary step between the original video sequence and the final composite image. This intermediary processing enhances subtle movements by analyzing phase information across multiple frames, making otherwise imperceptible motions visible and measurable, thereby resolving the contradiction between detecting subtle movements and maintaining processing complexity

Inventive Principle:
Principle #24Intermediary (Mediator)

2Adaptability or versatility

If multiple images from a sequence are combined to create a composite image, then the ability to capture optimal moments is improved, but the complexity of image manipulation increases

Engineering Contradiction:
Improveimage combination capabilityVSAvoidimage manipulation complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments the image manipulation process into distinct, manageable components: selecting a base image, identifying objects across multiple frames, determining their trajectories, and compositing them. This segmentation allows users to work with one object at a time rather than managing all images simultaneously, reducing the perceived complexity while maintaining high adaptability in combining multiple images

Inventive Principle:
Principle #1Segmentation

3Stability of the object's composition

If objects are repositioned along their movement trajectory, then the realism of the composite image is improved, but the precision required for trajectory determination increases

Engineering Contradiction:
Improverealism of composite imageVSAvoidtrajectory determination precision
Core Design Contradiction:
Stability of the object's compositionVSManufacturing precision

Solution Approach 1:

The system provides visual feedback by displaying the determined movement trajectory to the user during the editing process. This allows users to verify the accuracy of the automatically determined trajectory and make adjustments if needed, ensuring both realism in the final composite and transparency in the precision of trajectory determination

Inventive Principle:
Principle #23Feedback

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Enables users to create realistic composite images by repositioning objects along their trajectories and exaggerating subtle movements, providing a user-friendly interface for touch screen devices and avoiding unrealistic renderings by adhering to a deviation threshold.

Implementation Method 1

performance of a phase-based video processing algorithm to determine an exaggerated movement of an object in one of the images from the sequence that may otherwise only have a subtle movement

Methodology Applied
Scientific EffectPhase-based video motion processing:

Data Source

PatentEP3123448B1Image creation combining base image and repositioned object from a sequence of images
Publication Date: 2018.09.12 SONY GROUP CORP
  • EP3123448B1 patent drawingFigure 1A~1F
  • EP3123448B1 patent drawingFigure 2
  • EP3123448B1 patent drawingFigure 3

AI summary

According to one aspect of the present disclosure, a method implemented by a computing device is disclosed. Images from a sequence of images that depicts a scene are displayed on an electronic display. User input is received that selects an image from the sequence to be used as a base image. User input is also received that selects an object from the sequence of images. A movement trajectory of the selected object is determined from the sequence of images. The selected object is repositioned based on user input that drags the selected object along the determined movement trajectory from an initial position to a new position. A new image is created by combining the base image and the repositioned object.