• Home
  • Search
  • Refine pedestrian detections by referring to features in different ways
  • https://doi.org/10.1109/ivs.2017.7995754Copy DOI Icon

Refine pedestrian detections by referring to features in different ways

  • Jun 1, 2017
  • Jaemyung Lee +4 more
Show More
  • Abstract
  • Literature Map
  • References
  • Similar Papers
Abstract

The performance of object detection has been improved as the success of deep architectures. The main algorithm predominantly used for general detection is Faster R-CNN because of their high accuracy and fast inference time. In pedestrian detection, Region Proposal Network (RPN) itself which is used for region proposals in Faster R-CNN can be used as a pedestrian detector. Also, RPN even shows better performance than Faster R-CNN for pedestrian detection. However, RPN generates severe false positives such as high score backgrounds and double detections because it does not have downstream classifier. From this observations, we made a network to refine results generated from the RPN. Our Refinement Network refers to the feature maps of the RPN and trains the network to rescore severe false positives. Also, we found that different type of feature referencing method is crucial for improving performance. Our network showed better accuracy than RPN with almost same speed on Caltech Pedestrian Detection benchmark.

Similar Papers
  • Book Chapter
  • Citations959

Is Faster R-CNN Doing Well for Pedestrian Detection?

  • Jan 01, 2016
  • Liliang Zhang +3
  • Conference Article
  • Citations27

Deep convolutional neural networks for pedestrian detection with skip pooling

  • May 01, 2017
  • Jie Liu +4
  • Conference Article
  • Citations6

Pedestrian Detection Using Regional Proposal Network with Feature Fusion

  • Nov 01, 2018
  • Xiaogang Lv +3
  • Conference Article
  • Citations27

Pedestrian detection with dilated convolution, region proposal network and boosted decision trees

  • May 01, 2017
  • Jiqian Li +5
  • Research Article
  • Citations10

A Multi-strategy Region Proposal Network

  • Jun 28, 2018
  • Expert Systems with Applications
  • Yu-Peng Chen +3
  • PDF
  • Research Article

Deep Convolutional Neural Network for Pedestrian Detection with Multi-Levels Features Fusion

  • Jan 01, 2018
  • MATEC Web of Conferences
  • Danhua Li +4
  • Research Article
  • Citations55620

Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks.

  • Jun 06, 2016
  • IEEE Transactions on Pattern Analysis and Machine Intelligence
  • Shaoqing Ren +3
  • Research Article
  • Citations133

Faster ILOD: Incremental learning for object detectors based on faster RCNN

  • Sep 29, 2020
  • Pattern Recognition Letters
  • Can Peng +2
  • PDF
  • Research Article
  • Citations295

Vehicle Detection in Aerial Images Based on Region Convolutional Neural Networks and Hard Negative Example Mining

  • Feb 10, 2017
  • Sensors (Basel, Switzerland)
  • Tianyu Tang +4
  • Conference Article
  • Citations11

Research on Improved Pedestrian Detection Algorithm Based on Convolutional Neural Network

  • Jul 01, 2019
  • Jiachi Wang +3
  • Research Article

AI-Powered Waste Detection and Localization Using Region Proposal Network and CNN

  • Apr 21, 2025
  • INTERNATIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT
  • J Monicaa
  • Conference Article
  • Citations13

Seagrass Detection from Underwater Digital Images using Faster R-CNN with NASNet

  • Nov 01, 2021
  • Md Kislu Noman +3
  • Conference Article

Pedestrian Detection and Attribute Analysis Program Based on CNN

  • Oct 01, 2018
  • Zhang Junwei +3
  • Research Article
  • Citations29

Pedestrian Planar LiDAR Pose (PPLP) Network for Oriented Pedestrian Detection Based on Planar LiDAR and Monocular Images

  • Jan 10, 2020
  • IEEE Robotics and Automation Letters
  • Fan Bu +4
  • Book Chapter
  • Citations1

Text Localization in Scene Images Using Faster R-CNN with Double Region Proposal Networks

  • Jan 01, 2021
  • Pragya Hari +1
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.