Depth Map Enhancer Using Linear Optimizer for Low Power Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing depth sensors in electronic devices often generate incomplete or inaccurate initial depth maps, particularly in scenes with bright light sources or unideal reflective surfaces, which can impact the quality of depth information and hinder image processing applications.
Innovation Solution
A system comprising a scene depth sensor, an image generator, and a depth map enhancer that uses a linear optimizer to refine the initial depth map based on color images, identifying regions with inadequate depth information and generating estimated depth indicators using techniques such as super-pixel-level analysis and regularization terms to improve depth map accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If hardware depth sensors are used to generate initial depth maps, then depth information can be obtained, but the depth information may be incomplete or inaccurate in certain scenes
Solution Approach 1:
The patent introduces an intermediary processing system that includes a linear optimizer and multiple processing circuits (first processing circuit, second processing circuit, third processing circuit) acting as mediators between the depth sensor and the final depth map output. These intermediaries refine the initial depth map by identifying inadequate regions, generating estimated depth indicators, and combining results from multiple processing approaches to produce a complete and accurate refined depth map.
Solution Approach 2:
The system implements feedback mechanisms where the initial depth map is analyzed to identify regions with inadequate depth information, and this information feeds back into the processing circuits to generate corrective estimated depth indicators. The refined depth map is continuously improved by incorporating feedback from multiple processing circuits that evaluate different aspects of depth information quality.
2Measurement precision
If complex processing is applied to enhance depth maps, then depth information accuracy improves, but computational complexity and power consumption increase
Solution Approach 1:
The patent divides the depth map enhancement process into multiple segmented processing circuits, each handling specific tasks (first processing circuit for initial analysis, second processing circuit for depth indicator generation, third processing circuit for refinement). This segmentation allows parallel processing of different regions and aspects of the depth map, improving accuracy while distributing computational load efficiently to manage power consumption in electronic devices.
3Measurement precision
If regions with inadequate depth information are identified and refined, then depth map quality improves, but processing time increases
Solution Approach 1:
The system applies local quality enhancement by identifying specific regions with inadequate depth information and applying refined processing only to those local areas rather than the entire depth map. The processing circuits focus computational resources on regions needing improvement, maintaining high depth map quality while minimizing overall processing time through targeted local refinement.
Data Source
AI summary
An electronic device comprises circuitry implementing a depth map enhancer. The depth map enhancer obtains an initial depth map corresponding to a scene and an image of the scene. The depth map enhancer generates a refined depth map corresponding to the scene using an optimizer, the initial depth map and the image. The refined depth map includes estimated depth indicators corresponding to at least a first depth-information region, identified based at least in part on a first criterion, of the initial depth map. Input based on the refined depth map is provided to an image processing application.