Imaging System Depth Estimation Using Selective Sampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current imaging systems for depth estimation require significant manual input and computational strain, with existing methods lacking accuracy in detecting real lens positions and focus points, and are inefficient due to high computational load.
Innovation Solution
An imaging system that processes a sequence of images to generate an iteration map and depth map, using a subset of images to reduce computational load and improve accuracy, by determining dominant depths and combining maps to produce a highly accurate final depth map.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If automated methods are used for depth estimation, then manual input is reduced, but computational strain on computer systems increases
Solution Approach 1:
The patent divides the image processing task into segments by processing the image in multiple resolution levels (coarse to fine). The method first estimates depth at a lower resolution and then refines it at higher resolutions, reducing the computational burden at each stage while maintaining accuracy.
Solution Approach 2:
The patent uses a subset of images (8-20 images) instead of processing all 95 images from the full sequence. This partial action approach reduces computational strain while still achieving sufficient accuracy for depth estimation.
2Measurement precision
If more images are used for depth estimation, then measurement precision improves, but device complexity and computational load increase
Solution Approach 1:
The patent determines that a subset of 8-20 images is sufficient to achieve the desired measurement precision for depth estimation. This partial sampling approach avoids the need to process all 95 images while maintaining adequate accuracy, thereby reducing device complexity and computational load.
Solution Approach 2:
The patent segments the depth estimation process into multiple stages: initial coarse estimation using a subset of images, followed by refinement using additional images if necessary. This segmentation allows the system to achieve high precision without proportionally increasing computational complexity.
3Measurement precision
If real lens positions and focus points are detected accurately, then measurement precision improves, but computational strain increases
Solution Approach 1:
The patent segments the lens position detection into discrete steps by processing images at different resolution levels and using a subset of images. This segmentation reduces the computational strain required to accurately detect real lens positions and focus points.
Solution Approach 2:
The patent employs an iterative refinement process where initial depth estimates are used to guide subsequent processing stages. This feedback mechanism allows the system to converge on accurate lens position detection without requiring excessive computational resources from the outset.
Data Source
AI summary
A system and method of operation of an imaging system includes: a capture module for receiving an input sequence with an image pair, the image pair having a first image and a second image and for determining a first patch from the first image and a second patch from the second image; an iteration module for generating an iteration map from the first patch of the first image and the second patch of the second image; and a position module for generating a final depth map from the iteration map and for determining a focus position from the final depth map.