Publication:
End-to-End Rate-Distortion Optimization for Bi-Directional Learned Video Compression

dc.contributor.departmentDepartment of Electrical and Electronics Engineering
dc.contributor.kuauthorYılmaz, Mustafa Akın
dc.contributor.kuauthorTekalp, Ahmet Murat
dc.contributor.kuprofilePhD Student
dc.contributor.kuprofileFaculty Member
dc.contributor.yokidN/A
dc.contributor.yokid26207
dc.date.accessioned2024-11-09T23:00:07Z
dc.date.issued2020
dc.description.abstractConventional video compression methods employ a linear transform and block motion model, and the steps of motion estimation, mode and quantization parameter selection, and entropy coding are optimized individually due to combinatorial nature of the end-to-end optimization problem. Learned video compression allows end-to-end rate-distortion optimized training of all nonlinear modules, quantization parameter and entropy model simultaneously. While previous work on learned video compression considered training a sequential video codec based on end-to-end optimization of cost averaged over pairs of successive frames, it is well-known in conventional video compression that hierarchical, bi-directional coding outperforms sequential compression. In this paper, we propose for the first time end-to-end optimization of a hierarchical, bi-directional motion compensated learned codec by accumulating cost function over fixed-size groups of pictures (GOP). Experimental results show that the rate-distortion performance of our proposed learned bi-directional GOP coder outperforms the state-of-the-art end-to-end optimized learned sequential compression as expected.
dc.description.indexedbyWoS
dc.description.indexedbyScopus
dc.description.indexedbyPubMed
dc.description.openaccessYES
dc.description.sponsorshipTUBITAK project [217E033]
dc.description.sponsorshipTurkish Academy of Sciences (TUBA) This work was supported by TUBITAK project 217E033. A. Murat Tekalp also acknowledges support from Turkish Academy of Sciences (TUBA).
dc.identifier.doiN/A
dc.identifier.isbn978-1-7281-6395-6
dc.identifier.issn1522-4880
dc.identifier.scopus2-s2.0-85098622360
dc.identifier.urihttps://hdl.handle.net/20.500.14288/8005
dc.identifier.wos646178501083
dc.keywordsVideo compression
dc.keywordsDeep learning
dc.keywordsBi-directional motion compensation
dc.keywordsGroup of pictures
dc.keywordsEnd-to-end optimization
dc.languageEnglish
dc.publisherIeee
dc.source2020 Ieee International Conference On Image Processing (Icip)
dc.subjectDiagnostic imaging
dc.subjectPhotography
dc.titleEnd-to-End Rate-Distortion Optimization for Bi-Directional Learned Video Compression
dc.typeConference proceeding
dspace.entity.typePublication
local.contributor.authorid0000-0002-0795-8970
local.contributor.authorid0000-0003-1465-8121
local.contributor.kuauthorYılmaz, Mustafa Akın
local.contributor.kuauthorTekalp, Ahmet Murat
local.publication.orgunit1Graduate School of Sciences and Engineering
local.publication.orgunit1College of Engineering
local.publication.orgunit2Department of Electrical and Electronics Engineering
relation.isOrgUnitOfPublication21598063-a7c5-420d-91ba-0cc9b2db0ea0
relation.isOrgUnitOfPublication.latestForDiscovery21598063-a7c5-420d-91ba-0cc9b2db0ea0

Files