Depth Map Generation Using Microlens Angle Incidence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for generating depth maps from images captured by light-field cameras are time-consuming and require high-performance processors or reduced image frame processing, leading to increased hardware costs or compromised 3D image quality.
Innovation Solution
An image processing method and apparatus that determines depth levels based on the angle of incidence of light onto a microlens array, calculates a depth value for a reference object, estimates depth values for relative objects, and generates a depth map using these values, allowing for efficient depth map generation and refocusing capabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional weighted calculation methods are used to generate depth maps, then depth map generation is achieved, but processing time increases significantly
Solution Approach 1:
The patent segments the image into multiple depth levels based on the microlens array structure. Each microlens corresponds to a specific depth level, allowing the system to process depth information in discrete segments rather than calculating every pixel individually, thus reducing processing time while maintaining accuracy
Solution Approach 2:
The patent introduces an intermediary calculation approach by first determining depth levels from microlens angles and then calculating depth values only for reference objects, using these as intermediaries to estimate depths for all objects. This intermediary step significantly reduces the computational burden compared to direct pixel-by-pixel calculation
2Productivity
If a processor with better calculating performance is employed, then processing efficiency improves, but hardware costs increase
Solution Approach 1:
The patent replaces complex mechanical calculation systems with an optical-based depth estimation approach. By using the geometric relationship between microlens angles and object depths, the system substitutes computational complexity with optical geometry, achieving high processing efficiency without requiring powerful processors
Solution Approach 2:
The patent changes the parameters used for depth calculation from pixel-based weighted calculations to angle-based depth level determination. This parameter transformation enables linear time complexity processing, dramatically improving productivity while using standard hardware
3Productivity
If the number of image frames to be processed is reduced, then processing efficiency is maintained, but the quality of generated 3D images deteriorates
Solution Approach 1:
The patent applies partial action by calculating depth values only for reference objects (those at specific depth levels) rather than all objects in the image. This selective calculation maintains processing efficiency while the depth estimation for relative objects ensures sufficient 3D image quality without requiring processing of every frame at full resolution
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 improves the efficiency of depth map generation and enables cost-effective processing of multiple image frames while maintaining high-quality 3D image refocusing by directly calculating depth values for reference objects and estimating values for relative objects based on sensor locations.
Implementation Method 1
determining a depth level according to an angle of incidence in which light incident upon the object is projected onto the image sensor through the microlens array
Data Source
AI summary
In a method for generating a depth map for an image, an image processing apparatus is configured to: determine a depth level for each of at least two objects in the image according to an angle of incidence in which light incident upon the object is projected onto an image sensor of a light-field camera; calculate a depth value of the depth level associated with one of the objects; estimate a depth value for the depth level associated with another one of the objects; and generate a depth map according to the depth values. The depth value is estimated based on a distance between first and second locations on the image sensor, on which light incident upon the reference and relative objects are projected.