HM-net: A regression network for object center detection and tracking on wide area motion imagery

Motorcu, Hakkı; Ateş, Hasan Fehmi; Uğurdağ, Hasan Fatih; Güntürk, Bahadır Kürşat

HM-net: A regression network for object center detection and tracking on wide area motion imagery

dc.authorid	0000-0002-6842-1528
dc.authorid	0000-0003-0779-9620
dc.contributor.author	Motorcu, Hakkı
dc.contributor.author	Ateş, Hasan Fehmi
dc.contributor.author	Uğurdağ, Hasan Fatih
dc.contributor.author	Güntürk, Bahadır Kürşat
dc.date.accessioned	2022-01-13T07:14:39Z
dc.date.available	2022-01-13T07:14:39Z
dc.date.issued	2022
dc.department	İstanbul Medipol Üniversitesi, Mühendislik ve Doğa Bilimleri Fakültesi, Bilgisayar Mühendisliği Bölümü
dc.department	İstanbul Medipol Üniversitesi, Mühendislik ve Doğa Bilimleri Fakültesi, Elektrik ve Elektronik Mühendisliği Bölümü
dc.description.abstract	Wide Area Motion Imagery (WAMI) yields high resolution images with a large number of extremely small objects. Target objects have large spatial displacements throughout consecutive frames. This nature of WAMI images makes object tracking and detection challenging. In this paper, we present our deep neural network-based combined object detection and tracking model, namely, Heat Map Network (HM-Net). HM-Net is significantly faster than state-of-the-art frame differencing and background subtraction-based methods, without compromising detection and tracking performances. HM-Net follows object center-based joint detection and tracking paradigm. Simple heat map-based predictions support unlimited number of simultaneous detections. The proposed method uses two consecutive frames and the object detection heat map obtained from the previous frame as input, which helps HM-Net monitor spatio-temporal changes between frames and keep track of previously predicted objects. Although reuse of prior object detection heat map acts as a vital feedback-based memory element, it can lead to unintended surge of false positive detections. To increase robustness of the method against false positives and to eliminate low confidence detections, HM-Net employs novel feedback filters and advanced data augmentations. HM-Net outperforms state-of-the-art WAMI moving object detection and tracking methods on WPAFB dataset with its 96.2% F1 and 94.4% mAP detection scores, while achieving 61.8 % mAP tracking score on the same dataset. This performance corresponds to an improvement of 2.1% for F1, 6.1% for mAP scores on detection, and 9.5% for mAP score on tracking over state-of-the-art.
dc.identifier.citation	Motorcu, H., Ateş, H. F., Uğurdağ, H. F. ve Güntürk, B. K. (2022). HM-net: A regression network for object center detection and tracking on wide area motion imagery. IEEE Access, 10, 1346-1359. https://doi.org/10.1109/ACCESS.2021.3138980
dc.identifier.doi	10.1109/ACCESS.2021.3138980
dc.identifier.endpage	1359
dc.identifier.issn	2169-3536
dc.identifier.scopusquality	Q1
dc.identifier.startpage	1346
dc.identifier.uri	https://doi.org/10.1109/ACCESS.2021.3138980
dc.identifier.uri	https://hdl.handle.net/20.500.12511/8793
dc.identifier.volume	10
dc.identifier.wosquality	Q2
dc.indekslendigikaynak	Web of Science
dc.indekslendigikaynak	Scopus
dc.language.iso	en
dc.publisher	IEEE-Institute of Electrical and Electronics Engineers Inc.
dc.relation.ispartof	IEEE Access	en_US
dc.relation.publicationcategory	Makale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanı
dc.relation.tubitak	info:eu-repo/grantAgreement/TUBITAK/SOBAG/118E891
dc.rights	Attribution 4.0 International	*
dc.rights	info:eu-repo/semantics/openAccess
dc.rights.uri	https://creativecommons.org/licenses/by/4.0/	*
dc.subject	Tracking
dc.subject	Object Detection
dc.subject	Heating Systems
dc.subject	Target Tracking
dc.subject	Distortion
dc.subject	Convolutional Neural Networks
dc.subject	Cameras
dc.subject	Deep Neural Networks
dc.title	HM-net: A regression network for object center detection and tracking on wide area motion imagery
dc.type	Article