3D Video Creation Using Frame Selection and Transformation Matrix
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional techniques for converting 2D videos to 3D videos face challenges such as high calculation costs for estimating high-quality depth maps in real time, noise influence in depth map estimation, and time-consuming processes for accurate Structure from Motion (SFM) methods, leading to uncomfortable 3D experiences.
Innovation Solution
A three-dimensional video creating device that selects candidate frames with common areas and determines suitable partner frames based on criteria to create a 3D pair without relying on SFM or time-consuming depth map estimation, using a transformation matrix to convert frames into a stable and comfortable 3D video.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If depth map estimation is performed to convert 2D video to 3D video, then the quality of 3D image is improved, but the calculation cost and processing time increase significantly
Solution Approach 1:
The patent segments the video processing into two distinct paths: a fast path for flat images that skips depth map estimation entirely, and a detailed path for non-flat images that performs depth map estimation only when necessary. This segmentation resolves the contradiction by applying complex processing only where needed while maintaining simplicity for common cases.
Solution Approach 2:
The patent changes the parameter of processing intensity based on image characteristics. By detecting whether an image is flat or non-flat, the system dynamically adjusts the level of processing applied - using simple conversion for flat images and depth map-based conversion for non-flat images, thus balancing quality and computational cost.
2Manufacturing precision
If depth map estimation is performed to convert 2D video to 3D video, then the quality of 3D image is improved, but the processing time increases making real-time conversion difficult
Solution Approach 1:
The patent segments video frames into flat and non-flat categories, applying different processing speeds to each. Flat frames are processed rapidly without depth estimation, while non-flat frames receive detailed processing only when needed, achieving an average processing speed that enables real-time conversion.
Solution Approach 2:
The patent applies partial processing to flat images by skipping the depth map estimation step entirely, and applies full processing only to non-flat images. This partial action approach maintains acceptable 3D quality for most frames while enabling real-time processing throughput.
3Measurement precision
If SFM method is used for accurate 3D conversion, then the accuracy of 3D video is improved, but the processing time becomes too long for practical application
Solution Approach 1:
The patent segments the conversion process into a quick flat-image path and an accurate non-flat path. By detecting image characteristics first, the system avoids applying time-consuming SFM-like processing to flat images, reserving accurate methods only for frames where they are truly needed.
Solution Approach 2:
The patent performs preliminary classification of frames as flat or non-flat before applying conversion methods. This preliminary action allows the system to prepare and apply the appropriate processing level in advance, avoiding unnecessary computational steps and reducing overall processing time.
4Productivity
If conventional 2D to 3D conversion is applied to all frames, then the 3D video can be created, but camera shake effects are not reduced leading to uncomfortable viewing experience
Solution Approach 1:
The patent extracts and removes the stabilization step from the processing pipeline for flat images, applying it only selectively. By taking out the stabilization operation and applying it only where needed based on frame characteristics, the system reduces unnecessary processing while maintaining stability where required.
Data Source
AI summary
A three-dimensional video creating device (100) includes: a selection unit (123) which selects, from among frames constituting the 2D video, frames each of which has a common area whose proportion to the frame is greater than or equal to a predetermined value, as candidate three-dimensional partner frames that are candidate frames each constituting a three-dimensional image together with a target frame included in the frames constituting the 2D video; a determination unit (124) which determines, from among the candidate three-dimensional partner frames, a three-dimensional partner frame, based on the first criteria; and a three-dimensional pair creation unit (125) which creates a three-dimensional pair constituting the three-dimensional image corresponding to the target frame, using the target frame and the three-dimensional partner frame.