Region Proposal Network Score Boosting for Object Detection Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Deep learning approaches, particularly Faster R-CNN and R-FCN, face challenges in pedestrian detection due to high false positive rates, where background regions are incorrectly detected as foreground objects, limiting their effectiveness in applications like autonomous driving.
Innovation Solution
The proposed solution involves using a region proposal network (RPN) to assign confidence levels to regions of interest (RoIs), boosting the background score of the downstream classifier, and applying these scores in a softmax function to calculate final class probabilities, thereby reducing false positives and improving object detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If deep learning approaches (Faster R-CNN, R-FCN) are used for object detection, then general detection performance is improved, but false positive rate increases for pedestrian detection
Solution Approach 1:
The patent introduces an RPN score as an intermediary factor that mediates between the region proposal and the downstream classifier. This intermediary provides additional information about the likelihood of a region containing an object, which is then used to adjust the background scores of the classifier, thereby reducing false positives while maintaining detection accuracy.
Solution Approach 2:
The patent changes the parameter space by incorporating RPN scores into the classification process. Specifically, it modifies the background scores based on RPN confidence levels, effectively changing the probability parameters that the classifier uses to make detection decisions, which reduces false positive rates.
2Measurement precision
If RPN is used to identify regions of interest, then detection accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent applies preliminary action by using the RPN to pre-identify and score regions of interest before they are passed to the downstream classifier. This preliminary classification step filters and prioritizes regions, making the subsequent detection process more efficient and accurate without requiring complete re-processing of all image regions.
Data Source
AI summary
Apparatuses and methods of manufacturing same, systems, and methods for object detection using a region-based deep learning model are described. In one aspect, a method is provided, in which a region proposal network (RPN) is used to identify regions of interest (RoI) in an image by assigning a confidence levels, the assigned confidence levels of the RoIs are used to boost the background score assigned by the downstream classifier to each RoI, and the background scores are used in a softmax function to calculate the final class probabilities for each object class.