Hybrid IR Event Sensor for Real-Time 3D Scene Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sensor devices struggle to generate real-time three-dimensional models from two-dimensional images due to high computational requirements and the interference of visible light-based depth map generation, which is not suitable for applications like video conferencing and surveillance.
Innovation Solution
A sensor device integrating conventional video camera pixels (APS) with dynamic/event-based vision sensor (EVS) pixels sensitive to infrared light, allowing simultaneous capture of visible light video streams and additional infrared data for real-time depth map reconstruction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If multiple sensors are integrated into a single housing, then the device complexity is reduced and space is saved, but the risk of harmful factors affecting multiple sensors simultaneously increases
Solution Approach 1:
The patent divides the sensor housing into multiple separate sensor housings, each containing a single sensor. This segmentation isolates each sensor from harmful factors affecting others, while the control unit integrates the signal processing functions to maintain overall system simplicity.
Solution Approach 2:
The patent introduces an intermediary evaluation unit that centrally processes signals from multiple sensors. This intermediary layer separates the physical sensor elements from the signal processing functions, allowing sensors to be physically distributed while maintaining integrated control.
2Volume of moving object
If sensors are placed close together to reduce device size, then the device becomes more compact, but cross-sensitivity between sensors increases
Solution Approach 1:
The patent assigns each sensor its own dedicated housing, creating physical separation that prevents cross-sensitivity while maintaining compact overall device architecture through efficient spatial arrangement of the modular housing units.
Solution Approach 2:
The patent implements localized shielding and isolation measures within each sensor housing to prevent electromagnetic interference and cross-sensitivity, while allowing the overall device to maintain compact dimensions through optimized housing arrangement.
3Ease of manufacture
If a single housing is used for multiple sensors, then manufacturing is simplified, but signal interference between sensors occurs
Solution Approach 1:
The patent separates each sensor into its own housing unit, which simplifies individual sensor manufacturing and assembly while preventing signal interference. The modular design allows for standardized manufacturing processes that scale efficiently.
Solution Approach 2:
The patent introduces isolated signal transmission paths and an intermediary evaluation unit that processes signals from separated sensor housings. This intermediary layer prevents signal interference while maintaining manufacturing simplicity through standardized interface designs.
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 real-time conversion of two-dimensional images to three-dimensional models without disturbing the visible light video stream, providing enhanced data for applications like video conferencing and surveillance.
Implementation Method 1
a sensor housing (130) assigned to the sensor (120), wherein the sensor housing (130) is configured to shield the sensor (120) from harmful factors
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A sensor device (10) for observing a scene (S) comprises a plurality of pixels (51) each configured to receive light 5 from the scene and to perform photoelectric conversion to generate an electrical signal, event detection circuitry (20) that is configured to detect as event data intensity changes above a predetermined threshold of infrared light received by each of a first subset (S1) of the pixels (51), pixel signal generating circuitry (30) that is configured to generate pixel signals indicating intensity values of visible light received by each pixel (51) of a second subset (S2) of the pixels (51), and a control unit (40) that is configured to extract additional information from the event data detected within the received infrared light, which additional information differs from two-dimensional intensity information on the observed scene (S).