Virtual Point of View Image Generation System

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current methods for generating virtual point of view images from multiple camera captures lack efficiency in creating dynamic and flexible viewing experiences, particularly in capturing and displaying three-dimensional content with varied perspectives.

Innovation Solution

An image processing system that identifies and generates virtual point of view images based on multiple camera captures, allowing for dynamic perspective selection and display control, using a combination of shape estimation, image generation, and content creation units to produce three-dimensional digital content with multiple virtual points of view.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If virtual point of view images are generated from multiple camera captures, then viewing flexibility is improved, but processing time and system complexity increase

Engineering Contradiction:

Improveviewing flexibilityVSAvoidprocessing time

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The system pre-processes and stores multiple camera captures and their corresponding virtual point of view images before actual viewing. This allows rapid switching between different perspectives without performing complex image generation in real-time, thus reducing processing time while maintaining viewing flexibility.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

Instead of generating new virtual point of view images on-demand, the system uses pre-generated copies of virtual perspectives stored from multiple camera captures. This copying approach eliminates the need for time-consuming real-time rendering while preserving the flexibility to view from different angles.

Inventive Principle:

Principle #26Copying

2Ease of operation

If multiple virtual points of view are generated, then user interaction is improved, but device complexity increases

Engineering Contradiction:

Improveuser interactionVSAvoidsystem complexity

Core Design Contradiction:

Ease of operationVSDevice complexity

Solution Approach 1:

The system divides the complex task of generating multiple virtual perspectives into separate, manageable components. Each camera capture is independently processed to generate its corresponding virtual point of view image, which are then stored and retrieved as needed. This segmentation reduces the complexity of real-time processing while maintaining rich user interaction capabilities.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system introduces an intermediary storage layer that bridges the multiple camera captures and the final virtual point of view display. This intermediary stores pre-processed virtual perspectives, simplifying the overall system architecture by separating the complex image generation phase from the simpler retrieval and display phase during user interaction.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Productivity

If real-time perspective switching is enabled, then viewing experience is improved, but processing requirements increase

Engineering Contradiction:

Improveviewing experienceVSAvoidprocessing requirements

Core Design Contradiction:

ProductivityVSPower

Solution Approach 1:

The system performs all heavy image processing and virtual perspective generation in advance during a preliminary capture and processing phase. During actual viewing, the system simply retrieves pre-computed virtual point of view images from storage, dramatically reducing real-time processing requirements while enabling smooth perspective switching for enhanced viewing experience.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system maintains continuous availability of pre-generated virtual perspectives in storage, ensuring that perspective switching can occur immediately without interruption or delay. This continuity approach eliminates the need for continuous processing during viewing, reducing power requirements while preserving seamless user experience.

Inventive Principle:

Principle #20Continuity of useful action

Data Source

PatentUS20230353717A1Image processing system, image processing method, and storage medium

Publication Date: 2023.11.02 CANON KK

AI summary

It has been difficult to provide attractive digital content including a virtual point of view image and other images. The present disclosure is directed to providing an image processing system for providing a technique for displaying attractive digital content including a virtual point of view image and other images. Attractive digital content including a virtual point of view image and other images is displayed. The image processing system identifies a virtual point of view image and an image from a point of view different from a virtual point of view corresponding to the virtual point of view image from digital content of three-dimensional shape, and controls display of images corresponding to the virtual point of view image and the image from the point of view different from the virtual point of view corresponding to the virtual point of view image in a display area.