Depth of Field Simulation Using Unified Blur Parameter
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for simulating depth of field in computer-generated images are inefficient for real-time applications, requiring substantial processing power and often resulting in visual artifacts such as edge bleeding due to averaging blur values across pixels with different distance relationships.
Innovation Solution
A system and method that utilize a single blur parameter to represent both the magnitude of blur and relative distance for each object or pixel, employing a spatially variant filter with weight values based on blur magnitude and distance, to simulate depth of field without averaging blur values, thus preserving distance relationships and reducing processing intensity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional depth of field simulation methods are used, then depth of field effects can be simulated, but processing requirements and computational complexity increase substantially
Solution Approach 1:
The patent changes the parameter representation by using a single blur parameter that encodes both magnitude and relative distance information. This parameter transformation allows the system to maintain depth of field simulation accuracy while reducing computational complexity, as the unified parameter eliminates the need for separate blur magnitude and distance calculations for each pixel
Solution Approach 2:
The patent extracts and eliminates the harmful averaging operation from the depth of field simulation process. By removing the averaging step that causes edge bleeding artifacts, the system maintains simulation accuracy without requiring additional complex processing to correct the artifacts, thus reducing overall processing requirements
2Productivity
If blur values are averaged across pixels to simplify processing, then processing intensity is reduced, but visual artifacts such as edge bleeding occur
Solution Approach 1:
The patent removes the averaging operation from the processing pipeline entirely. By extracting this harmful step, the system avoids generating edge bleeding artifacts while maintaining processing efficiency through the use of the unified blur parameter that enables direct depth of field simulation without iterative averaging
Solution Approach 2:
Instead of averaging blur values and then trying to correct artifacts, the patent inverts the approach by using a single blur parameter that directly represents the depth of field effect without requiring averaging. This reversal of the processing logic eliminates the artifact generation step while maintaining computational efficiency
3Measurement precision
If multiple parameters are used to represent blur magnitude and distance separately, then accuracy is improved, but processing complexity increases
Solution Approach 1:
The patent merges the blur magnitude and relative distance information into a single unified blur parameter. This combination allows the system to maintain precise depth representation through the encoded parameter while significantly reducing processing complexity by eliminating the need to handle and coordinate multiple separate parameters for each pixel
Solution Approach 2:
The single blur parameter serves multiple functions simultaneously: it encodes both the magnitude of blur and the relative distance information needed for depth of field simulation. This multi-functionality eliminates the need for separate parameter storage and processing, reducing computational complexity while maintaining measurement precision
Data Source
AI summary
A system for simulating depth of field in a computer graphics display is provided according to an exemplary embodiment. The system comprises an image generation circuit configured to generate an image representing a plurality of objects for display on the computer graphics display. The image generation circuit is configured to generate the image using a single blur parameter for each of the plurality of objects that represents both a magnitude of blur and a relative distance with respect to a lens focal distance being simulated by the image generation circuit.