Camera Depth Map Creation via Focus Variation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Surveillance cameras, especially monocular stationary cameras, face challenges in obtaining depth information for scenes, which is essential for robust object tracking and video content analysis, as existing methods are either time-consuming or require expensive 3D sensors.

Innovation Solution

A method that creates a depth map by varying the focus settings of a camera to determine depth information in selected sub-areas of the detection area, using image gradients to select and control partial areas, and combining these to generate a precise depth map, which can be used to enhance object tracking and video surveillance applications.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If depth information is obtained by moving the camera around the room to capture from different angles, then depth map can be determined, but the process is time-consuming and complex

Engineering Contradiction:
Improvedepth information accuracyVSAvoidtime for capturing depth information
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The camera's field of view is divided into multiple sub-areas, and depth information is determined separately for each sub-area by varying focus settings. This segmentation allows depth mapping to be performed on small sections rather than requiring comprehensive scene scanning from multiple camera positions.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The focus setting of the camera is varied to determine depth information for different sub-areas. By changing the focus parameter, the camera can capture depth information at different focal planes, enabling depth map creation without physically moving the camera through the scene.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If depth information is obtained using 3D sensors, then accurate depth mapping is achieved, but the system becomes expensive

Engineering Contradiction:
Improvedepth information accuracyVSAvoidcamera system cost
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The camera system uses its own existing focus mechanism to determine depth information, rather than requiring external 3D sensors. The camera's autofocus capability is repurposed to generate depth maps, allowing the system to serve its own depth measurement needs using components already present in the camera.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The focus setting parameter of the existing camera is varied to extract depth information from standard 2D images. This parameter change approach enables a conventional camera to function as a depth-sensing device without adding expensive 3D sensing hardware.

Inventive Principle:
Principle #35Parameter changes

3Area of stationary object

If the entire detection range is scanned for depth information, then complete depth coverage is achieved, but the process becomes time-consuming

Engineering Contradiction:
Improvedepth map coverage areaVSAvoidtime for depth mapping
Core Design Contradiction:
Area of stationary objectVSLoss of time

Solution Approach 1:

The detection range is divided into multiple sub-areas, and depth information is determined for each sub-area separately by varying focus settings. This segmentation enables efficient processing of large areas by breaking them into manageable sections that can be processed independently and quickly.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Depth information is determined for selected sub-areas rather than uniformly across the entire detection range. This partial action approach focuses computational resources on specific regions of interest, reducing overall processing time while still providing useful depth coverage for surveillance purposes.

Inventive Principle:
Principle #16Partial or excessive action

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

This method allows for precise depth mapping without the need for extensive scene scanning or expensive 3D sensors, providing valuable depth information for improved object tracking and video content analysis, and can be integrated into existing camera systems for enhanced surveillance capabilities.

Implementation Method 1

depth information is determined by varying the focus setting of the camera for each sub-area

Methodology Applied
Scientific EffectFocus: Focusing

Data Source

PatentEP3185213B1Method for creating a depth map by means of a camera
Publication Date: 2021.04.14 ROBERT BOSCH GMBH
  • EP3185213B1 patent drawingFigure 1
  • EP3185213B1 patent drawingFigure 2a~2b
  • EP3185213B1 patent drawingFigure 3~4

AI summary

The invention relates to a method for creating a depth map of at least one selected area (210) of a detection area (200) of a camera (100), wherein several different sub-areas (220, 221, 222) of the at least one selected area (210) are controlled by means of the camera (100), wherein depth information from the respective sub-area (220, 221, 222) is determined for each of the sub-areas (220, 221, 222) by varying a focus setting of the camera (100), and wherein the depth map (400) is created taking into account the depth information of the sub-areas (220, 221, 222), as well as such a camera (100).