Publication:
Video frame prediction via deep learning

dc.contributor.departmentDepartment of Electrical and Electronics Engineering
dc.contributor.departmentDepartment of Electrical and Electronics Engineering
dc.contributor.kuauthorTekalp, Ahmet Murat
dc.contributor.kuauthorYılmaz, Mustafa Akın
dc.contributor.kuprofileFaculty Member
dc.contributor.schoolcollegeinstituteCollege of Engineering
dc.contributor.yokid26207
dc.contributor.yokidN/A
dc.date.accessioned2024-11-09T12:39:36Z
dc.date.issued2020
dc.description.abstractThis paper provides new results over our previous work presented in ICIP 2019 on the performance of learned frame prediction architectures and associated training methods. More specifically, we show that using an end-to-end residual connection in the fully convolutional neural network (FCNN) provides improved performance. In order to provide comparative results, we trained a residual FCNN, a convolutional RNN (CRNN), and a convolutional long-short term memory (CLSTM) network for next frame prediction using the mean square loss. We performed both stateless and stateful training for recurrent networks. Experimental results show that the residual FCNN architecture performs the best in terms of peak signal to noise ratio (PSNR) at the expense of higher training and test (inference) computational complexity. The CRNN can be stably and efficiently trained using the stateful truncated backpropagation through time procedure, and requires an order of magnitude less inference runtime to achieve an acceptable performance in near real-time. / Bu bildiri, ICIP 2019’da sundugumuz öğrenilmiş video çerçeve öngörü mimarileri ve egitim yöntemleri karşılaştırılması çalışmamız üzerine yeni sonuçlar vermektedir. Daha spesifik olarak, tamamen evrisimsel sinir agında (FCNN) uçtan uca kalıntı bağlantısı kullanarak öngörü performansında gelişim sağlandığını gösteriyoruz. Sonuçları karşılaştırmak için, ortalama kare kaybını kullanarak sonraki kareyi öngörmek için kalıntı baglantılı FCNN, evrisimsel bir RNN (CRNN) ve evrisimsel uzun-kısa süreli bellek (CLSTM) agı eğittik. Yinelemeli ağlar için durum bilgisi taşımayan (stateless) ve taşıyan (stateful) egitim gerçekleştirdik. Deneysel sonuçlar, kalıntı baglantılı FCNN mimarisinin yüksek egitim ve test (çıkarım) hesaplama karmaşıklığı pahasına PSNR açısından en iyi performansı verdigini göstermektedir. CRNN, durum bilgisi taşıyan zamanda geri yayılım yoluyla kararlı olarak egitilebilmekte ve kabul edilebilir bir performansla gerçek-zamanlıya yakın çerçeve öngörüsü yapabilmektedir.
dc.description.fulltextYES
dc.description.indexedbyWoS
dc.description.indexedbyScopus
dc.description.openaccessYES
dc.description.publisherscopeInternational
dc.description.sponsoredbyTubitakEuTÜBİTAK
dc.description.sponsorshipScientific and Technological Research Council of Turkey (TÜBİTAK)
dc.description.sponsorshipTurkish Academy of Sciences (TUBA)
dc.description.versionAuthor's final manuscript
dc.formatpdf
dc.identifier.doi10.1109/SIU49456.2020.9302047
dc.identifier.embargoNO
dc.identifier.filenameinventorynoIR03846
dc.identifier.isbn978-1-7281-7206-4
dc.identifier.issn2165-0608
dc.identifier.linkhttps://doi.org/10.1109/SIU49456.2020.9302047
dc.identifier.quartileN/A
dc.identifier.scopus2-s2.0-85100304312
dc.identifier.urihttps://hdl.handle.net/20.500.14288/2112
dc.identifier.wos653136100021
dc.keywordsFrame prediction
dc.keywordsDeep learning
dc.keywordsRecurrent network architectures
dc.keywordsStateful training
dc.keywordsConvolutional network architectures
dc.languageTurkish
dc.publisherInstitute of Electrical and Electronics Engineers (IEEE)
dc.relation.grantno2.18E+37
dc.relation.urihttp://cdm21054.contentdm.oclc.org/cdm/ref/collection/IR/id/10714
dc.source2020 28th Signal Processing and Communications Applications Conference (SIU)
dc.subjectElectrical and electronic engineering
dc.subjectTelecommunications
dc.titleVideo frame prediction via deep learning
dc.title.alternativeDerin ögrenme ile video çerçeve öngörüsü
dc.typeConference proceeding
dspace.entity.typePublication
local.contributor.authorid0000-0003-1465-8121
local.contributor.authoridN/A
local.contributor.kuauthorTekalp, Ahmet Murat
local.contributor.kuauthorYılmaz, Mustafa Akın
relation.isOrgUnitOfPublication21598063-a7c5-420d-91ba-0cc9b2db0ea0
relation.isOrgUnitOfPublication.latestForDiscovery21598063-a7c5-420d-91ba-0cc9b2db0ea0

Files

Original bundle

Now showing 1 - 1 of 1
Thumbnail Image
Name:
10714.pdf
Size:
414.74 KB
Format:
Adobe Portable Document Format