A Fully Convolutional Encoder-Decoder Network for Moving Object Segmentation

Anil Turker, Ender M. Eksioglu

Araştırma sonucu: Kitap/Rapor/Konferans Bildirisinde BölümKonferans katkısıbilirkişi

2 Atıf (Scopus)

Özet

Moving object segmentation (MOS) is one of the important and well-studied computer vision problems. It is used in applications such as video surveillance systems, human tracking, self-driving cars, and video compression. Traditional approaches solve this problem by using hand-crafted features and then modeling the background by using these features. Convolutional Neural Networks (CNNs), on the other hand, have proven to be more powerful than traditional methods in extracting features. In this work, a hybrid system is presented that contains flux tensors together with 3D CNN, enhancing the performance of the algorithm on the unseen videos. 3D CNN can extract spatial and temporal features, thus exploiting motion information between adjacent frames. Motion entropy feature maps extracted by 3D CNN and the output of the flux tensor are jointly fed into an encoder-decoder network. ChangeDetection 2014 dataset is used for both training and test stages. Training and test videos are selected separately, and the networks are tested on unseen videos. Our proposed network gives promising segmentation results, which are competitive with existing methods.

Orijinal dilİngilizce
Ana bilgisayar yayını başlığı16th International Conference on INnovations in Intelligent SysTems and Applications, INISTA 2022
EditörlerRichard Chbeir, Tulay Yildirim, Ladjel Bellatreche, Yannis Manolopoulos, Apostolos Papadopoulos, Karam Bou Chaaya
YayınlayanInstitute of Electrical and Electronics Engineers Inc.
ISBN (Elektronik)9781665498104
DOI'lar
Yayın durumuYayınlandı - 2022
Etkinlik16th International Conference on INnovations in Intelligent SysTems and Applications, INISTA 2022 - Biarritz, France
Süre: 8 Ağu 202212 Ağu 2022

Yayın serisi

Adı16th International Conference on INnovations in Intelligent SysTems and Applications, INISTA 2022

???event.eventtypes.event.conference???

???event.eventtypes.event.conference???16th International Conference on INnovations in Intelligent SysTems and Applications, INISTA 2022
Ülke/BölgeFrance
ŞehirBiarritz
Periyot8/08/2212/08/22

Bibliyografik not

Publisher Copyright:
© 2022 IEEE.

Parmak izi

A Fully Convolutional Encoder-Decoder Network for Moving Object Segmentation' araştırma başlıklarına git. Birlikte benzersiz bir parmak izi oluştururlar.

Alıntı Yap