Light Field Stereoscopy for Per-Pixel Disparity Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for generating stereoscopic content lack granular control over binocular disparity, particularly in real-world images, leading to limitations in depth perception and image distortion.
Innovation Solution
The method involves generating stereoscopic content from a 3D representation of a light field by determining a point within the light field that matches a desired binocular disparity for each pixel, allowing for per-pixel control of disparity through multi-perspective imaging.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If remapping image disparities and re-synthesizing new images is used to control binocular disparity, then disparity control is improved, but image distortion and loss of information occur due to hole filling requirements
Solution Approach 1:
The patent transitions from 2D image remapping to 3D light field sampling, adding a spatial dimension to the disparity control process. By sampling the light field at multiple perspectives and synthesizing views through ray tracing in 3D space, the system achieves disparity control without the hole-filling problems of 2D remapping, as genuine light ray information is available from the multi-perspective capture.
2Adaptability or versatility
If accurate disparity computation and hole filling are performed for disparity remapping, then disparity control is improved, but device complexity and processing difficulty increase
Solution Approach 1:
The patent performs preliminary action by capturing the light field with multi-perspective cameras before disparity control is needed. The light field data structure pre-contains light ray information from multiple viewpoints, so when disparity control is required, the system can directly sample and synthesize views without performing complex real-time disparity computation and hole filling, significantly reducing processing complexity.
3Ease of manufacture
If a fixed camera baseline is used for stereoscopic imaging, then device complexity is reduced, but adaptability to different scenes and viewing preferences is limited
Solution Approach 1:
The patent implements dynamics by making the effective camera baseline adjustable through software processing rather than fixed hardware configuration. The light field data allows dynamic selection of virtual camera positions and baselines, enabling the system to adapt to different scenes, viewing conditions, and user preferences without changing the physical camera setup, thus achieving both simplicity and adaptability.
Data Source
AI summary
Methods and systems for generating stereoscopic content with granular control over binocular disparity based on multi-perspective imaging from representations of light fields are provided. A reference image, a three-dimensional (“3D”) representation of a light field corresponding to the reference image, and a goal disparity image that indicates a goal binocular disparity for one or more pixels of the reference image may be received. For each pixel of an output image corresponding to the reference image, a point within the light field that is a closest match for the goal binocular disparity of a corresponding pixel of the goal disparity image may be determined. A stereoscopic image pair including the reference image and the output image may be generated.