Infrared Imaging Foreground Background Separation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing depth imaging technologies rely on specialized cameras that consume more energy, require higher computational power, and are costly, limiting their adoption in consumer electronics due to form-factor constraints.
Innovation Solution
The method uses infrared imaging to distinguish foreground from background in a scene, even with unstructured or diffuse infrared illumination, employing a combined visible light (VL) and infrared (IR) imaging approach that does not require purpose-built 3D cameras, allowing for foreground modification and background replacement or blurring, and can be implemented on mobile devices like smartphones.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If specialized depth cameras are used for depth imaging, then depth imaging capability is improved, but energy consumption increases and device complexity increases
Solution Approach 1:
The patent combines visible light imaging and infrared imaging into a single system, using the infrared camera to provide depth information that enhances the visible light camera's depth imaging capability. This merging allows the system to achieve depth imaging without requiring separate specialized depth cameras, thereby reducing energy consumption and device complexity while maintaining improved depth measurement precision.
Solution Approach 2:
The infrared camera serves multiple functions: it provides depth information for depth imaging, enables foreground-background separation, and works in conjunction with the visible light camera to create a multi-functional imaging system. This multi-functionality eliminates the need for dedicated specialized depth cameras, reducing overall system complexity and energy consumption while maintaining depth imaging capability.
2Measurement precision
If specialized depth cameras are used for depth imaging, then depth imaging capability is improved, but device complexity increases
Solution Approach 1:
The patent merges visible light imaging and infrared imaging into a unified system where the infrared camera's depth information enhances the visible light camera's capabilities. This integration achieves depth imaging functionality without adding separate specialized depth cameras, thereby reducing device complexity while maintaining improved depth measurement precision.
Solution Approach 2:
The infrared camera performs multiple functions including depth imaging, foreground-background separation, and collaborative imaging with the visible light camera. This multi-functionality consolidates what would otherwise require multiple specialized devices, reducing overall system complexity while maintaining enhanced depth imaging capability.
3Measurement precision
If specialized depth cameras are used for depth imaging, then depth imaging capability is improved, but cost increases
Solution Approach 1:
The patent combines visible light and infrared imaging capabilities in a single system, leveraging the infrared camera to provide depth information that enhances overall depth imaging capability. This approach avoids the need for expensive specialized depth cameras, achieving improved depth measurement precision at a lower cost.
Solution Approach 2:
The infrared camera serves multiple purposes including depth imaging, foreground-background separation, and collaborative imaging, replacing what would otherwise require multiple specialized expensive devices. This multi-functionality reduces the overall system cost while maintaining improved depth imaging capability.
4Measurement precision
If infrared imaging is used with unstructured or diffuse illumination, then foreground-background separation is achieved, but illumination control becomes more difficult
Solution Approach 1:
The system utilizes naturally occurring or ambient infrared illumination without requiring structured or controlled infrared light sources. The infrared camera captures depth information from the unstructured or diffuse infrared illumination present in the environment, enabling foreground-background separation without the complexity of illumination control mechanisms.
Solution Approach 2:
The patent changes the illumination parameter from structured or controlled infrared illumination to unstructured or diffuse infrared illumination. This parameter change simplifies the illumination system while maintaining the ability to achieve foreground-background separation through the infrared camera's depth-sensing capability.
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 approach enables efficient depth imaging on ubiquitous 2D cameras, reducing energy consumption and costs while providing effective foreground-background separation for applications like real-time video communication and natural user interfaces, without the need for specialized hardware.
Implementation Method 1
obtain one or more infrared (IR) images captured via an IR camera
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An initial candidate foreground region is identified within an infrared image that includes pixels exhibiting infrared intensity values within a pre-defined range. A depth of surfaces within the initial candidate foreground region is estimated based on infrared intensity values the pixels of the initial candidate foreground region. The initial candidate foreground region is expanded to an expanded candidate foreground region based on a body-model estimate. The body model estimate is seeded with one or more of the initial candidate foreground region, the depth of surfaces, and/or a face of a human subject identified by facial recognition. Each pixel of the infrared image is identified as either a foreground pixel or a background pixel based on a distance of that pixel relative to the expanded candidate foreground region. Pixels identified as background pixels may be modified within a corresponding visible light image.