Please use this identifier to cite or link to this item: https://hdl.handle.net/11147/7559
Title: Training CNNs with image patches for object localisation
Authors: Orhan, Semih
Baştanlar, Yalın
Orhan, Semih
Baştanlar, Yalın
Izmir Institute of Technology. Computer Engineering
Keywords: Object detection
Computer vision
Object recognition
Convolutional Neural Networks
Animal species
Issue Date: Apr-2018
Publisher: Institution of Engineering and Technology
Source: Orhan, S., and Baştanlar, Y. (2018). Training CNNs with image patches for object localisation. Electronics Letters, 54(7), 424-426. doi:10.1049/el.2017.4725
Abstract: Recently, convolutional neural networks (CNNs) have shown great performance in different problems of computer vision including object detection and localisation. A novel training approach is proposed for CNNs to localise some animal species whose bodies have distinctive patterns such as leopards and zebras. To learn characteristic patterns, small patches which are taken from different body parts of animals are used to train models. To find object location, in a test image, all locations are visited in a sliding window fashion. Crops are fed into trained CNN and their classification scores are combined into a heat map. Later on, heat maps are converted to bounding box estimates for varying confidence scores. The localisation performance of the patch-based training approach is compared with Faster R-CNN – a state-of-the-art CNN-based object detection and localisation method. Experimental results reveal that the patch-based training outperforms Faster R-CNN, especially for classes with distinctive patterns.
URI: https://doi.org/10.1049/el.2017.4725
https://hdl.handle.net/11147/7559
ISSN: 0013-5194
0013-5194
Appears in Collections:Computer Engineering / Bilgisayar Mühendisliği
Scopus İndeksli Yayınlar Koleksiyonu / Scopus Indexed Publications Collection
WoS İndeksli Yayınlar Koleksiyonu / WoS Indexed Publications Collection

Files in This Item:
File Description SizeFormat 
orhan2018.pdfMakale (Article)589.75 kBAdobe PDFThumbnail
View/Open
Show full item record

CORE Recommender

SCOPUSTM   
Citations

7
checked on Sep 18, 2021

WEB OF SCIENCETM
Citations

7
checked on Sep 18, 2021

Page view(s)

36
checked on Sep 21, 2021

Download(s)

36
checked on Sep 21, 2021

Google ScholarTM

Check

Altmetric


Items in GCRIS Repository are protected by copyright, with all rights reserved, unless otherwise indicated.