Accession Number:

AD1013692

Title:

Keypoint Density-Based Region Proposal for Fine-Grained Object Detection and Classification Using Regions with Convolutional Neural Network Features

Descriptive Note:

Technical Report

Corporate Author:

KNEXUS RESEARCH CORP NATIONAL HARBOR MD NATIONAL HARBOR

Personal Author(s):

• Turner,J. T.
• Gupta,Kalyan
• Morris,Brendan
• Aha,David W.

Report Date:

2015-12-15

Pagination or Media Count:

9.0

Abstract:

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.

Descriptors:

• algorithms
• image processing
• digital data
• artificial neural networks
• detection
• feature extraction
• images
• artificial intelligence
• classification

Subject Categories:

Distribution Statement:

APPROVED FOR PUBLIC RELEASE