Computational Shallow Depth of Field Generation via Region-Based Blur

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Digital cameras, including those in smartphones, often lack the capability to produce a sufficiently shallow depth of field effect due to inappropriate f-number ratios, necessitating a method to artificially generate this effect.

Innovation Solution

A method and system that calculate the difference in focus or object distance between regions of interest and non-interest regions in an image, determine the degree of blur based on this difference, and apply it to generate a blurred image, simulating a shallow depth of field effect.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Illumination intensity

If digital cameras use standard f-number ratios, then the camera can capture sufficient light and maintain good exposure, but the depth of field remains too deep and cannot produce the desired shallow bokeh effect

Engineering Contradiction:
Improvelight capture capabilityVSAvoidshallow depth of field effect
Core Design Contradiction:
Illumination intensityVSEase of manufacture

Solution Approach 1:

The image is segmented into multiple depth layers (foreground, midground, background) based on focus distance calculations. Each layer is independently processed with different blur amounts, allowing selective application of shallow depth of field effect to specific regions while preserving detail in the focused subject area.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different blur strengths are applied to different spatial regions of the image based on their depth relative to the focus point. The foreground and background regions receive stronger blur, while the midground (subject) remains sharp, creating a localized shallow depth of field effect that varies across the image rather than being uniform.

Inventive Principle:
Principle #3Local quality

2Ease of manufacture

If the camera lens is designed with a lower f-number to achieve shallow depth of field, then the bokeh effect is improved, but the camera cannot capture sufficient light and exposure quality deteriorates

Engineering Contradiction:
Improveshallow depth of field effectVSAvoidlight capture capability
Core Design Contradiction:
Ease of manufactureVSIllumination intensity

Solution Approach 1:

The system performs preliminary depth mapping and focus distance calculation during image capture, then applies the shallow depth of field effect as a post-processing operation. This allows the camera to capture the image with adequate light using a higher f-number, then synthetically create the shallow depth of field appearance afterward, avoiding the light loss that would occur with a physically lower f-number.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent creates a synthetic copy of the shallow depth of field effect through computational methods rather than relying on the physical optical properties of the lens. By calculating expected blur based on depth information and applying it algorithmically, the system reproduces the appearance of shallow depth of field without the optical constraints that would normally require a low f-number.

Inventive Principle:
Principle #26Copying

3Device complexity

If computational blur is applied uniformly to the entire image, then processing is simplified, but the shallow depth of field effect becomes unnatural and loses depth information

Engineering Contradiction:
Improveprocessing complexityVSAvoiddepth information accuracy
Core Design Contradiction:
Device complexityVSManufacturing precision

Solution Approach 1:

The image is divided into multiple depth-based segments (foreground, midground, background) with distinct blur amounts applied to each. This segmentation preserves the natural progression of depth information, with closer elements remaining sharper and distant elements becoming progressively blurrier, maintaining the authentic appearance of optical shallow depth of field while still using computational methods.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The blur amount is dynamically adjusted for each pixel or region based on its calculated depth relative to the focus point. Rather than applying a static uniform blur, the system varies the blur strength continuously across the image according to depth information, creating a natural transition from sharp to blurred regions that accurately represents the three-dimensional scene structure.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP3573331B1Method and system for generating shallow depth of field effect
Publication Date: 2021.08.04 BLACKBERRY LTD
  • EP3573331B1 patent drawingFigure 1
  • EP3573331B1 patent drawingFigure 2
  • EP3573331B1 patent drawingFigure 3

AI summary

Methods and systems for generating a shallow depth of field effect for a digitally captured image are provided. At least one region of interest (ROI) and at least one non-interest region are defined in the captured image. A difference in focus or object distance is calculated between the ROI and each non-interest region. A degree of blur is applied to each non-interest region, based on the calculated difference in focus or object distance.