3D Virtual Space Synthesis for 2D Image Depth
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing techniques require 3D-to-2D conversion for processing 3D images, leading to complex processes and the need for additional components, and are limited by the absence of 3D space information in general electronic devices without 3D capturing cameras.
Innovation Solution
An electronic device with a processor that sets virtual planes in a 3D virtual position based on 2D images, creates a 3D virtual space, inserts 3D objects, and generates 2D images by synthesizing the 3D objects with the 2D images, allowing for the provision of a 3D effect to 2D images without a 3D capturing camera.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If 3D-to-2D conversion is used to process 3D images, then image processing can be performed, but the process becomes complicated and additional components are required
Solution Approach 1:
Instead of converting 3D images to 2D for processing, the patent inverts the approach by generating 3D space information from 2D images. The processor creates virtual planes and 3D virtual spaces based on 2D image data, then inserts 3D objects into this generated 3D space to produce synthesized images with 3D effects. This inversion eliminates the need for complex 3D-to-2D conversion processes and additional components.
Solution Approach 2:
The patent creates a virtual copy of 3D space information by generating virtual planes and 3D virtual spaces from 2D image data. This virtual 3D space serves as a substitute for actual 3D captured images, allowing 3D image processing to be performed on standard 2D images without requiring expensive 3D capturing cameras or complex conversion processes.
2Measurement precision
If a 3D capturing camera is used to identify 3D space information, then 3D image processing can be performed, but the camera is expensive and not widely used
Solution Approach 1:
The patent creates a virtual copy of 3D space information by generating virtual planes and 3D virtual spaces from 2D image data. This virtual 3D space serves as a substitute for actual 3D captured images, allowing 3D image processing to be performed on standard 2D images without requiring expensive 3D capturing cameras.
Solution Approach 2:
The patent replaces the mechanical/optical system of 3D capturing cameras with a computational approach. Instead of using physical 3D sensors to capture and identify 3D space information, the system uses software-based image processing to generate 3D virtual spaces from 2D images, substituting hardware-based 3D capture with software-based 3D generation.
3Adaptability or versatility
If general image processing techniques are used without 3D capturing camera, then device compatibility is maintained, but 3D space information cannot be identified
Solution Approach 1:
The patent transitions from 2D image space to 3D virtual space by generating virtual planes and 3D virtual spaces from 2D image data. This dimensionality change allows the system to work with standard 2D images while creating and processing 3D virtual environments, enabling 3D space information identification without requiring 3D capturing cameras.
Solution Approach 2:
The patent changes the parameter representation from 2D image coordinates to 3D virtual space parameters. By generating virtual planes with 3D coordinates and creating 3D virtual spaces, the system transforms 2D image data into 3D parameter space, enabling 3D space information identification while maintaining compatibility with general image processing techniques.
Data Source
AI summary
An electronic device includes a memory, a display and a processor configured to set at least one virtual plane in a three-dimensional (3D) virtual position based on at least a portion of at least one two-dimensional (2D) first image displayed on the display, set a 3D virtual space comprising the at least one virtual plane and the at least one 2D first image, insert a 3D object into the 3D virtual space, and generate at least one 2D second image by synthesizing the 3D object inserted into the 3D virtual space and the at least one 2D first image.