Depth Map Generation Using Multi-Focal Plane Variance Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for generating depth maps in 3D computer graphics often require multiple cameras or a structured light source, which can be impractical in certain situations.

Innovation Solution

A system and method using an image capture sensor to capture images at multiple focal planes, calculating variance values, determining peak variance values, and generating depth maps based on these values to determine the depth of objects in a 3D scene without the need for multiple cameras or structured light.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If multiple cameras or structured light source are used to generate depth map, then measurement precision is improved, but device complexity increases

Engineering Contradiction:
Improvedepth map accuracyVSAvoidnumber of cameras
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent transitions from using multiple cameras in different spatial positions to capturing images at multiple focal planes (depth dimension). By focusing at different depths and analyzing variance in sharpness across these focal planes, the system extracts depth information without requiring multiple cameras, thus resolving the contradiction between measurement precision and device complexity

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The patent changes the focal length parameter of a single camera across multiple captures to encode depth information. By varying the focus distance and measuring image variance at each focal plane, the system derives depth maps using parameter changes rather than adding more cameras, thereby maintaining accuracy while reducing device complexity

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If multiple cameras are used to generate depth map, then depth information accuracy is improved, but ease of operation deteriorates

Engineering Contradiction:
Improvedepth information accuracyVSAvoidsystem setup complexity
Core Design Contradiction:
Measurement precisionVSEase of operation

Solution Approach 1:

The patent makes a single camera perform multiple functions: capturing color images and extracting depth information through multi-focal plane imaging. The same camera sensor and lens are used for both photometric and depth measurement purposes by varying focus, eliminating the need for separate depth cameras and simplifying system operation

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent merges the depth sensing function with the standard color imaging camera. By combining multi-focal plane captures with variance analysis, the system integrates depth measurement capabilities into a single camera unit, reducing the number of components and simplifying system setup and operation

Inventive Principle:
Principle #5Merging (Combining)

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

Enables the generation of accurate depth maps using a single image capture sensor and processor, facilitating applications such as object manipulation and visualization in 3D environments.

Implementation Method 1

an image capture sensor to capture a plurality of images of the object at a plurality of different focal planes

Methodology Applied
Scientific EffectFocusing: Focusing

Implementation Method 2

calculate a plurality of variance values for a plurality of image locations for each image captured by the image capture sensor

Methodology Applied
Scientific EffectVariance calculation:

Data Source

PatentEP3252716B1Depth map from multi-focal plane images
Publication Date: 2022.03.16 RECOGNITION ROBOTICS INC
  • EP3252716B1 patent drawingFigure 1
  • EP3252716B1 patent drawingFigure 2
  • EP3252716B1 patent drawingFigure 3

AI summary

A system for generating a depth map for an object in a three-dimensional (3D) scene includes an image capture sensor and a processor. The image capture sensor is configured to capture a plurality of images of the object at a plurality of different focal planes. The processor is configured to calculate a plurality of variance values for a plurality of image locations for each image captured by the image capture sensor. The processor is also configured to determine a peak variance value for the plurality of image locations based on the calculated variance values associated with the same image location for each of the plurality of images of the object at the plurality of different focal planes. The processor is also configured to generate the depth map for the object based on the determined peak variance value for each image location and the plurality of different focal planes.