Canvas View Generation Using Optical Flow for VR
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for generating canvas views in virtual reality are slow and require manual stitching, and they struggle with discrepancies in camera views due to differences in brightness or color between camera images.
Innovation Solution
The system generates a canvas view by creating a synthetic view based on optical flow between camera views, shifting and blending them to approximate a desired view, using a combination of mappings to associate regions of the canvas view with camera views, and applying these mappings to produce a panoramic view that can be displayed in virtual reality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If manual stitching techniques are used to generate canvas views, then the quality of view alignment can be maintained, but the generation speed becomes slow and productivity decreases
Solution Approach 1:
The patent replaces manual stitching operations with an automated optical flow-based system. The optical flow algorithm automatically computes pixel correspondences between camera views and generates the canvas view without human intervention, thereby maintaining alignment quality while dramatically improving generation speed and productivity.
Solution Approach 2:
The system performs self-alignment by automatically computing optical flow fields and generating canvas views without requiring manual stitching. The algorithm independently handles view registration and blending, enabling the system to serve itself and eliminating the need for manual operations.
2Productivity
If direct blending of camera views is performed without optical flow, then the generation process is simple and fast, but discrepancies in brightness and color between views cannot be handled
Solution Approach 1:
The patent introduces optical flow as an intermediary mechanism between camera views and the final canvas view. The optical flow field serves as a mediator that computes precise pixel correspondences and guides the blending process, enabling the system to handle brightness and color discrepancies while maintaining generation speed.
3Manufacturing precision
If optical flow is used to align camera views, then brightness and color discrepancies can be handled, but the computational complexity increases
Solution Approach 1:
The patent applies optical flow computation selectively and efficiently, computing flow fields only where needed for view alignment rather than performing exhaustive computations across the entire image space. This partial action approach maintains high alignment accuracy while reducing unnecessary computational complexity.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enables efficient generation of high-quality canvas views that provide a 3D stereoscopic effect by accurately combining and aligning camera views, reducing manual input and handling discrepancies in brightness and color.
Implementation Method 1
An optical flow can be used associating pixels between camera views represented as a set of optical flow vectors each associating two or more corresponding pixels
Data Source
AI summary
A canvas generation system generates a canvas view of a scene based on a set of original camera views depicting the scene, for example to recreate a scene in virtual reality. Canvas views can be generated based on a set of synthetic views generated from a set of original camera views. Synthetic views can be generated, for example, by shifting and blending relevant original camera views based on an optical flow across multiple original camera views. An optical flow can be generated using an iterative method which individually optimizes the optical flow vector for each pixel of a camera view and propagates changes in the optical flow to neighboring optical flow vectors.