Keypoint Density-based Region Proposal for Fine-Grained Object Detection and Classification using Regions with Convolutional Neural Network Features
This addresses the problem of real-time performance in fine-grained object detection for applications like robotics or surveillance, though it is incremental as it improves an existing region proposal technique.
The paper tackles the slow response time of regional CNNs for real-time fine-grained object detection and classification by proposing a Keypoint Density-based Region Proposal (KDRP) method, which speeds up detection by 100% compared to selective search without losing accuracy.
Although recent advances in regional Convolutional Neural Networks (CNNs) enable them to outperform conventional techniques on standard object detection and classification tasks, their response time is still slow for real-time performance. To address this issue, we propose a method for region proposal as an alternative to selective search, which is used in current state-of-the art object detection algorithms. We evaluate our Keypoint Density-based Region Proposal (KDRP) approach and show that it speeds up detection and classification on fine-grained tasks by 100% versus the existing selective search region proposal technique without compromising classification accuracy. KDRP makes the application of CNNs to real-time detection and classification feasible.