ROI-Based Object Recognition for Faster Autonomous Driving Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current object recognition systems for autonomous driving face inefficiencies in processing large images, leading to increased inference times due to the need to analyze all regions of an input image, which hampers the speed and accuracy of object detection.
Innovation Solution
The proposed method employs a neural network architecture that extracts a region of interest (ROI) from an input image, determines the number of object candidate regions based on the ROI's size, and uses a region-based convolutional neural network (R-CNN) with a region proposal network (RPN) to recognize objects, thereby reducing the number of candidate regions and enhancing processing speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all regions of an input image are analyzed for object detection, then detection coverage is improved, but inference time increases
Solution Approach 1:
The input image is divided into multiple regions of interest (ROIs) based on semantic segmentation, allowing the system to focus object detection only on relevant regions rather than processing the entire image, thus reducing inference time while maintaining detection coverage
Solution Approach 2:
Regions of interest are extracted from the input image based on segmentation results, and object detection is performed only on these extracted ROIs, eliminating the need to analyze irrelevant regions and significantly reducing processing time
2Measurement precision
If the number of object candidate regions is increased, then detection accuracy is improved, but processing complexity increases
Solution Approach 1:
Semantic segmentation is performed as a preliminary step to identify and extract regions of interest before object detection, which pre-filters the search space and reduces the number of candidate regions that need to be evaluated, thereby reducing processing complexity while maintaining detection accuracy
Data Source
AI summary
A method and apparatus for recognizing an object are provided, including extracting a feature from an input image and generating a feature map in a neural network. In parallel with the generating of the feature map, a region of interest (ROI) corresponding to an object of interest is extracted from the input image, and a number of object candidate regions used to detect the object of interest is determined based on a size of the ROI. The object of interest is recognized from the ROI based on the number of object candidate regions in the neural network.