Shallow Depth of Field Generation via Computational Blur
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current camera lenses with adjustable diaphragms and post-processing methods struggle to achieve a natural and accurate shallow depth of field effect, as they rely on general perception for blur estimation, leading to inaccuracies and unnatural images.
Innovation Solution
A three-dimensional imaging system using left and right lenses captures images, calculates disparity between corresponding pixels, estimates depth information based on lens focus and pitch, and blurs pixels accordingly to generate a more authentic shallow depth of field effect.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If lenses with large apertures are used to enhance the blurring of long distance objects, then the shallow depth of field effect is improved, but the camera becomes bulky and expensive
Solution Approach 1:
The patent creates a virtual copy of the optical blurring effect through computational methods. Instead of using physical lenses with large apertures, the system captures multiple images with different focus settings and processes them to generate a synthetic shallow depth of field effect, thereby avoiding the need for bulky optical hardware while achieving the desired visual outcome
Solution Approach 2:
The patent replaces the mechanical/optical system (large aperture lenses) with a computational system. By using image processing algorithms to simulate depth of field effects based on focus stack analysis, the system substitutes physical optical components with digital processing, eliminating the need for bulky lens hardware
2Device complexity
If post-processing methods are used to create shallow depth of field effect, then the camera structure remains simple, but the blur level estimation is inaccurate and causes discontinuous or unnatural images
Solution Approach 1:
The patent implements a feedback mechanism where the system captures multiple images at different focus settings, analyzes the focus quality of each image to determine depth information, and uses this feedback to guide the blending process. This iterative approach allows the system to accurately estimate blur levels by comparing actual focus measurements rather than relying on general perception algorithms
Solution Approach 2:
The patent performs preliminary actions by capturing a stack of focused images at different depth planes before the final image generation. This pre-capture of multiple focus states provides the necessary data foundation for accurate depth analysis and blur level estimation, enabling natural-looking shallow depth of field effects without requiring complex real-time processing
3Device complexity
If general perception is used to determine blur level of objects, then the processing is simple, but the depth information is inaccurate for objects outside focal range
Solution Approach 1:
The system uses feedback from actual image analysis to determine depth information. By examining the focus quality metrics of captured images at different focus settings, the system receives feedback about which objects are in focus and which are out of focus, allowing it to accurately estimate depth and blur levels based on measured data rather than perceptual assumptions
Data Source
AI summary
A method and an apparatus for generating an image with shallow depth of field, suitable for a three-dimensional imaging system having a left lens and a right lens, are provided. First, the left lens and the right lens are used to capture a left-eye image and a right-eye image. Next, a disparity between each of a plurality of corresponding pixels in the left and right-eye images is calculated. Then, the depth information of each pixel is estimated according to a focus of the left and right lens, a lens pitch between the left lens and the right lens, and the calculated disparity of each pixel. Finally, the pixels in the left-eye image and the right-eye image are blurred according to the focus of the left lens and the right lens and the estimated depth information of each pixel, so as to obtain the image with shallow depth of field.