Object Deformation via Polyhedron Apex Vector Decomposition

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing object manipulation techniques, such as skeleton animation and GUI-based hand finger detection, face challenges in reducing arithmetic processing load and achieving stable deformation and smooth manipulation of objects on moving images, due to complex coordinate system calculations and low joint constraint forces.

Innovation Solution

The method involves setting a polyhedron with a given point as the center, dividing the space using vectors, discriminating the apex location, storing decomposition coefficients, and deforming the object based on deformation manipulation, which reduces the need for weighted orthogonal coordinate system calculations and eliminates the need for joint consideration, allowing for stable deformation and manipulation of objects on moving images.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If skeleton animation with skinning is used to represent characters in a real manner, then the expression realism is improved, but the arithmetic processing complexity increases due to weighted addition of orthogonal coordinate systems and definition of joint elasticities

Engineering Contradiction:
Improveexpression realismVSAvoidarithmetic processing complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent divides the object into multiple apexes (vertices) that can be independently manipulated. Each apex is controlled by its own transformation matrix, allowing localized deformation without complex global coordinate system transformations. This segmentation approach reduces arithmetic complexity while maintaining realistic expression.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent changes the representation parameters from traditional skeleton bone coordinate systems to a polyhedron-based apex transformation system. Instead of using weighted addition of orthogonal coordinate systems, the invention uses direct transformation matrices applied to apexes, significantly reducing the arithmetic processing required while achieving similar or better deformation results.

Inventive Principle:
Principle #35Parameter changes

2Manufacturing precision

If physically modeling bones with coordinate systems and joint elasticities is used, then the deformation accuracy is improved, but the joint constraint force becomes insufficient and instability occurs due to small number of links

Engineering Contradiction:
Improvedeformation accuracyVSAvoidjoint stability
Core Design Contradiction:
Manufacturing precisionVSReliability

Solution Approach 1:

The patent segments the object into multiple apexes connected through a polyhedron structure with transformation matrices. This creates a more robust deformation model where each apex can be independently controlled, providing better constraint stability compared to traditional bone-link models with few links.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements dynamic transformation matrices that can adaptively adjust the deformation of each apex based on input manipulation data. This dynamic approach allows the system to maintain stability during deformation by adjusting transformation parameters in real-time, overcoming the limitations of static bone models with fixed joint constraints.

Inventive Principle:
Principle #15Dynamics

3Ease of operation

If GUI member is displayed on hand finger part image as still image, then the GUI operation functionality is improved, but the smooth manipulation following hand movement on moving images becomes difficult

Engineering Contradiction:
ImproveGUI operation functionalityVSAvoidsmooth manipulation speed
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The patent applies dynamic transformation matrices to the displayed object that update in synchronization with hand movement detection. Instead of treating the object as a static still image, the system continuously adjusts the object's position and deformation based on real-time hand gesture data, enabling smooth manipulation following on moving images while maintaining full GUI operation functionality.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentUS10080963B2Object manipulation method, object manipulation program, and information processing apparatus
Publication Date: 2018.09.25 SONY INTERACTIVE ENTERTAINMENT LLC
  • US10080963B2 patent drawing
  • US10080963B2 patent drawing
  • US10080963B2 patent drawing

AI summary

In an object (OB) manipulation method of manipulating an object (OB) displayed on a display apparatus for deformation, a control apparatus configured to control the display apparatus executes a step of setting a polyhedron (POL) with a given point (P0) in the object (OB) being a center; a step of dividing a space inside the polyhedron (POL) by a vector going from the center (P0) of the polyhedron (POL) to apexes (P1 through P12) of the polyhedron (POL); a step of discriminating in which of the divided spaces an apex of the object (OB) is included; a step of storing a decomposition coefficient of the apex (POB) of the object (OB) by three vectors making up the space including the apex (POB) of the object (OB); and a step of deforming the object (OB) in accordance with the deformation manipulation acted on the polyhedron (POL) and displaying the deformed object (OB).