3D Semantic Parking Space Detection for Smooth Parking Assistance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing parking assistance systems fail to accurately detect parking spaces using depth information, leading to difficulty in implementing smooth parking assistance, especially in environments where depth recognition is challenging.
Innovation Solution
An information processing device and method that utilizes 3D semantic segmentation to generate images with depth data and class information, enabling the detection of available parking spaces.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If parking space detection is performed without depth information, then the system complexity is reduced, but the detection precision and reliability deteriorate in environments where depth recognition is challenging
Solution Approach 1:
The patent transitions from 2D image-based parking space detection to 3D detection by incorporating depth information. The 3D semantic segmentation model processes both image data and depth data to generate a 3D segmentation image, enabling the system to accurately detect parking spaces in various environments including those where depth recognition is challenging, thereby resolving the contradiction between system complexity and detection precision.
2Device complexity
If traditional 2D image-based detection is used, then the device complexity is low, but the reliability of parking space detection deteriorates in challenging environments
Solution Approach 1:
The patent combines multiple data types (image data and depth data) to create a composite 3D semantic segmentation model. This composite approach integrates information from different sensors and processing methods, enabling reliable parking space detection in challenging environments while maintaining reasonable system complexity through unified model processing.
3Speed
If depth information is not considered in parking space detection, then the processing speed is faster, but the productivity of parking assistance deteriorates due to unnecessary searching time
Solution Approach 1:
The patent performs preliminary 3D semantic segmentation processing to identify and segment parking spaces, drivable areas, and obstacles before the actual parking assistance operation. This preliminary action with depth information enables the system to quickly determine valid parking spaces, reducing unnecessary searching time and improving overall parking assistance efficiency despite the additional processing required.
Data Source
AI summary
To implement quick and smooth parking assistance close to a human sense without being affected by an environment of a parking space. A 3D semantic segmentation image having a plurality of pixel units, each pixel unit of the plurality of pixel units including the depth data and class information, are generated, an available parking space is searched for on the basis of the 3D semantic segmentation image, and a path to the searched parking space is planned and a vehicle is controlled. The present disclosure can be applied to a parking assistance system.