Sensor Information Conversion for Occlusion-Free Object Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless communication systems face limitations in accurately sharing sensor information among devices, particularly in crowded environments, leading to shadow areas and difficulties in detecting multiple objects due to interference, overlap, or occlusion, which hampers effective object detection and positioning.
Innovation Solution
A method and device for converting and transmitting sensor information by fusing data from multiple sensors to minimize shadow areas and enhance object detection, using techniques such as 3D rendering and information conversion from the perspective of the receiver, allowing for more accurate position and size information transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If sensor information is transmitted from the transmitter's perspective, then the transmission process is simple, but the receiver has difficulty detecting objects accurately due to shadow areas and occlusion
Solution Approach 1:
The patent inverts the traditional sensor information transmission approach by converting data from the transmitter's perspective to the receiver's perspective. Instead of sending raw sensor data and letting the receiver interpret it, the system performs viewpoint transformation to present objects as they would appear to the receiving device, thereby improving detection accuracy while maintaining transmission simplicity
Solution Approach 2:
The patent introduces an intermediary processing step that transforms sensor information between different reference frames. This intermediary conversion process acts as a mediator that reconciles the transmitter's sensor data with the receiver's detection needs, eliminating shadow areas and occlusion issues without complicating the overall system architecture
2Area of stationary object
If multiple sensors are used to detect objects, then object detection coverage is improved, but shadow areas and interference increase making detection more difficult
Solution Approach 1:
The patent merges sensor information from multiple sources and time points into a unified viewpoint representation. By combining data from multiple sensors and performing viewpoint conversion, the system creates a consolidated object detection result that eliminates shadow areas and interference, making detection easier while maintaining comprehensive coverage
Solution Approach 2:
The patent adds a temporal dimension to the detection process by utilizing sensor data from multiple time points. This multi-temporal approach, combined with viewpoint transformation, allows the system to reconstruct object positions and eliminate shadow areas that would be present in single-moment snapshots, thereby reducing detection difficulty while expanding coverage
3Productivity
If sensor data is transmitted without conversion, then transmission efficiency is high, but the receiver cannot accurately determine object position and size
Solution Approach 1:
The patent performs viewpoint conversion as a preliminary action before transmission. By pre-processing the sensor data to convert it to the receiver's perspective, the system ensures that the transmitted information is already in the correct format for accurate position and size determination, eliminating the need for complex post-reception processing while maintaining transmission efficiency
Data Source
AI summary
Provided are a method by which a first device performs wireless communication, and a device supporting same. The method may comprise the steps of: obtaining first state information of an object on the basis of at least one of information acquired through one or more sensors connected to the first device or information received from one or more devices; converting the first state information into second state information from the perspective of a second device; and transmitting a message including the second state information to the second device.