Restricted Depth of Field Depth Map Synthesis for Array Camera
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing camera systems face challenges in efficiently synthesizing images with varying depth estimation precision and spatial resolution, particularly when capturing light fields with multiple cameras, as they struggle to balance computational resources and image quality while maintaining a realistic depth of field effect.
Innovation Solution
The system generates a restricted depth of field depth map using a processor and image processing pipeline, which determines a desired focal plane distance and range of distances for image rendering, providing higher precision for pixels within the restricted depth of field and lower precision for those outside, allowing for efficient rendering of images with a synthetic depth of field effect.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If uniform high precision depth estimation is applied to all pixels, then image synthesis quality is improved, but computational requirements increase significantly
Solution Approach 1:
The patent applies different depth estimation precision levels to different regions of the image based on their depth characteristics. Pixels corresponding to objects within the restricted depth of field receive high precision depth estimation, while pixels for objects outside this range receive lower precision estimation. This local differentiation maintains image synthesis quality for critical regions while significantly reducing overall computational requirements.
2Manufacturing precision
If full resolution depth maps are generated from multiple camera views, then image synthesis accuracy is improved, but processing time increases
Solution Approach 1:
The patent segments the depth map generation process into two distinct stages: a first stage that generates a low-resolution depth map quickly from multiple camera views, and a second stage that generates high-resolution depth estimates only for pixels within the restricted depth of field by processing individual images. This segmentation allows the system to achieve high synthesis accuracy for critical regions while maintaining fast processing times through the initial quick low-resolution pass.
3Adaptability or versatility
If depth of field is restricted to a specific range, then image realism is improved, but depth estimation precision is reduced for out-of-focus regions
Solution Approach 1:
The patent applies partial precision to depth estimation based on the functional requirements of different regions. For pixels within the restricted depth of field (the critical region for image realism), high precision depth estimation is applied. For pixels outside this range, where depth information is less critical for achieving the desired depth of field effect, lower precision estimation is applied. This partial application of high precision only where needed resolves the contradiction between depth of field control and measurement precision.
Data Source
AI summary
Systems and methods are described for generating restricted depth of field depth maps. In one embodiment, an image processing pipeline application configures a processor to: determine a desired focal plane distance and a range of distances corresponding to a restricted depth of field for an image rendered from a reference viewpoint; generate a restricted depth of field depth map from the reference viewpoint using the set of images captured from different viewpoints, where depth estimation precision is higher for pixels with depth estimates within the range of distances corresponding to the restricted depth of field and lower for pixels with depth estimates outside of the range of distances corresponding to the restricted depth of field; and render a restricted depth of field image from the reference viewpoint using the set of images captured from different viewpoints and the restricted depth of field depth map.