Automatic Object Scaling Using Depth Map Ratios

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing image editing techniques fail to produce natural-looking results when moving or copying objects between images, as they often result in improper scaling and occlusion issues, leading to unrealistic appearances.

Innovation Solution

The method utilizes depth maps to automatically adjust the scale and occlusion of objects when moved or copied, determining the appropriate scaling factor and depth values by considering the object's interaction with its surroundings, ensuring proper size and positioning relative to its new environment.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If automatic scaling is not applied when moving/copying objects, then the object size remains unchanged, but the object appears unnatural and improperly sized in the new location

Engineering Contradiction:
Improveobject size accuracyVSAvoidmanual scaling effort
Core Design Contradiction:
Manufacturing precisionVSEase of operation

Solution Approach 1:

The system automatically determines the scaling factor by comparing depth values at the source and target locations, enabling the object to self-adjust its size without manual intervention. The processor calculates the ratio between depth values and applies this scaling factor automatically during the paste operation.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent changes the size parameter of the object dynamically based on depth map information. By retrieving depth values at different locations and calculating their ratio, the system transforms the object's dimensional parameters to match the perspective requirements of the target location.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If depth map processing is implemented for automatic scaling, then object placement realism is improved, but processing complexity increases

Engineering Contradiction:
Improverealism of object placementVSAvoidprocessing system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The depth map is generated in advance before the object moving/copying operation. By having the depth information pre-computed and available, the system can quickly retrieve depth values at specific locations during the paste operation without performing complex real-time depth calculations.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The depth map serves as an intermediary data structure that mediates between the original image and the object pasting operation. It provides easy access to depth information at any location, simplifying the scaling calculation process compared to performing direct geometric computations.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Measurement precision

If manual scaling adjustment is required, then precise control over object size is achieved, but the process becomes time-consuming and less efficient

Engineering Contradiction:
Improvescaling accuracyVSAvoidimage editing efficiency
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The system uses depth map feedback to automatically determine the appropriate scaling factor. By comparing the depth value at the source location with the depth value at the target location, the system computes the scaling factor that maintains realistic proportions, eliminating the need for manual trial-and-adjustment processes.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3040943B1Automatic scaling of objects based on depth map for image editing
Publication Date: 2020.06.17 SONY GROUP CORP
  • EP3040943B1 patent drawingFigure 1
  • EP3040943B1 patent drawingFigure 2A~2B
  • EP3040943B1 patent drawingFigure 3~4A

AI summary

Automatic scaling of image objects (52) being moved or copied to a target object location (54) is presented. An original image and its depth map are received within which the user selects an original object (52) and a target location (54) for the object to be moved or copied. A center of scaling (O') is found and a base location (B') selected at which depth value is to be obtained at the target location (54) for use in scaling. Target object is then scaled in response to a ratio between original object depth and target object depth. An occlusion check is made, after which non-occluded pixels are pasted to the target location to complete the move/copy which is in proper proportion and accounts for surrounding structures.