Academic Journal
Προβολή σε EDS

Effective methods in video coding and inpainting

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος: Effective methods in video coding and inpainting
Συνεισφορές: Chen, Zhe (author.), Cham, Wai-Kuen (thesis advisor.), Blu, Thierry (thesis advisor.), Chinese University of Hong Kong Graduate School. Division of Electronic Engineering. (degree granting institution.)
Έτος έκδοσης: 2019
Συλλογή: The Chinese University of Hong Kong: CUHK Digital Repository / 香港中文大學數碼典藏
Θεματικοί όροι: Video compression, Digital video--Editing--Data processing, Coding theory, TK6680.5 .C479 2019eb
Περιγραφή: Ph.D. ; In recent years, high-definition videos and video-based applications become more and more popular, boosting the demand of advanced video processing techniques. This thesis proposes effective methods in video coding and inpainting based on the information redundancy in natural videos. ; This thesis proposes integer approximations of the Discrete Cosine Transforms (DCTs) to compactly represent the signals generated in video coding. These approximations include order-8, 16, 32 and 64 Integer Cosine Transforms (ICTs) with fully factorizable structures. The proposed ICTs achieve similar coding performance as the state-of-art methods on the High Efficiency Video Coding (HEVC) test model while requiring lower complexity. In addition, it is observed that order-64 ICTs achieve around 1% average bitrate reduction under low-bitrate configurations. ; This thesis also investigates the scenarios of video inpainting, where the original signals are missing and have to be inferred from the related valid signals. It proposes a fast video inpainting method to tackle dynamic backgrounds. Based on the analysis of the displacements between similar patches, a novel label pruning scheme is proposed to dramatically accelerate the 3D Markov Random Field (MRF) based inpainting framework and make it applicable to high-resolution dynamic backgrounds. Experiment results show that this method is faster than the state-of-art methods and achieves promising results on videos with challenging background dynamisms. ; In addition to background inpainting, a foreground inpainting method is proposed to tackle periodically-moving foreground objects under variable illumination. By assuming linear color change during the damaged interval, a novel color compensation scheme is proposed to reduce the color inconsistency between the inpainted objects and valid ones. Experiment results demonstrate the effectiveness of this method. ; 近年來，逐漸流行的高清視頻及基於視頻的各種應用催生了對高效視頻處理技術的需求。基於自然視頻中的信息冗餘，本論文提出了用于視頻編碼及修友的一系列有效方法。 ...
Τύπος εγγράφου: text
Περιγραφή αρχείου: electronic resource; remote; 1 online resource (ii-xiv, 89 leaves) : illustrations (some color); computer; online resource
Γλώσσα: English
Chinese
Relation: cuhk:2327162; local: ETD920200822; local: AAI27662455; local: 991039842707603407
Διαθεσιμότητα: https://julac.hosted.exlibrisgroup.com/primo-explore/search?query=addsrcrid,exact,991039842707603407,AND&tab=default_tab&search_scope=All&vid=CUHK&mode=advanced&lang=en_US
https://repository.lib.cuhk.edu.hk/en/item/cuhk-2327162
Rights: Use of this resource is governed by the terms and conditions of the Creative Commons "Attribution-NonCommercial-NoDerivatives 4.0 International" License (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Αριθμός Καταχώρησης: edsbas.7EDE2BF
Βάση Δεδομένων: BASE
FullText Text:
  Availability: 0
CustomLinks:
  – Url: https://julac.hosted.exlibrisgroup.com/primo-explore/search?query=addsrcrid,exact,991039842707603407,AND&tab=default_tab&search_scope=All&vid=CUHK&mode=advanced&lang=en_US#
    Name: EDS - BASE (ns324271)
    Category: fullText
    Text: View record from BASE
Header DbId: edsbas
DbLabel: BASE
An: edsbas.7EDE2BF
RelevancyScore: 883
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 883.215454101563
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Effective methods in video coding and inpainting
– Name: Author
  Label: Contributors
  Group: Au
  Data: Chen, Zhe (author.)<br />Cham, Wai-Kuen (thesis advisor.)<br />Blu, Thierry (thesis advisor.)<br />Chinese University of Hong Kong Graduate School. Division of Electronic Engineering. (degree granting institution.)
– Name: DatePubCY
  Label: Publication Year
  Group: Date
  Data: 2019
– Name: Subset
  Label: Collection
  Group: HoldingsInfo
  Data: The Chinese University of Hong Kong: CUHK Digital Repository / 香港中文大學數碼典藏
– Name: Subject
  Label: Subject Terms
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Video+compression%22">Video compression</searchLink><br /><searchLink fieldCode="DE" term="%22Digital+video--Editing--Data+processing%22">Digital video--Editing--Data processing</searchLink><br /><searchLink fieldCode="DE" term="%22Coding+theory%22">Coding theory</searchLink><br /><searchLink fieldCode="DE" term="%22TK6680%2E5+%2EC479+2019eb%22">TK6680.5 .C479 2019eb</searchLink>
– Name: Abstract
  Label: Description
  Group: Ab
  Data: Ph.D. ; In recent years, high-definition videos and video-based applications become more and more popular, boosting the demand of advanced video processing techniques. This thesis proposes effective methods in video coding and inpainting based on the information redundancy in natural videos. ; This thesis proposes integer approximations of the Discrete Cosine Transforms (DCTs) to compactly represent the signals generated in video coding. These approximations include order-8, 16, 32 and 64 Integer Cosine Transforms (ICTs) with fully factorizable structures. The proposed ICTs achieve similar coding performance as the state-of-art methods on the High Efficiency Video Coding (HEVC) test model while requiring lower complexity. In addition, it is observed that order-64 ICTs achieve around 1% average bitrate reduction under low-bitrate configurations. ; This thesis also investigates the scenarios of video inpainting, where the original signals are missing and have to be inferred from the related valid signals. It proposes a fast video inpainting method to tackle dynamic backgrounds. Based on the analysis of the displacements between similar patches, a novel label pruning scheme is proposed to dramatically accelerate the 3D Markov Random Field (MRF) based inpainting framework and make it applicable to high-resolution dynamic backgrounds. Experiment results show that this method is faster than the state-of-art methods and achieves promising results on videos with challenging background dynamisms. ; In addition to background inpainting, a foreground inpainting method is proposed to tackle periodically-moving foreground objects under variable illumination. By assuming linear color change during the damaged interval, a novel color compensation scheme is proposed to reduce the color inconsistency between the inpainted objects and valid ones. Experiment results demonstrate the effectiveness of this method. ; 近年來，逐漸流行的高清視頻及基於視頻的各種應用催生了對高效視頻處理技術的需求。基於自然視頻中的信息冗餘，本論文提出了用于視頻編碼及修友的一系列有效方法。 ...
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: text
– Name: Format
  Label: File Description
  Group: SrcInfo
  Data: electronic resource; remote; 1 online resource (ii-xiv, 89 leaves) : illustrations (some color); computer; online resource
– Name: Language
  Label: Language
  Group: Lang
  Data: English<br />Chinese
– Name: NoteTitleSource
  Label: Relation
  Group: SrcInfo
  Data: cuhk:2327162; local: ETD920200822; local: AAI27662455; local: 991039842707603407
– Name: URL
  Label: Availability
  Group: URL
  Data: https://julac.hosted.exlibrisgroup.com/primo-explore/search?query=addsrcrid,exact,991039842707603407,AND&tab=default_tab&search_scope=All&vid=CUHK&mode=advanced&lang=en_US<br />https://repository.lib.cuhk.edu.hk/en/item/cuhk-2327162
– Name: Copyright
  Label: Rights
  Group: Cpyrght
  Data: Use of this resource is governed by the terms and conditions of the Creative Commons "Attribution-NonCommercial-NoDerivatives 4.0 International" License (http://creativecommons.org/licenses/by-nc-nd/4.0/)
– Name: AN
  Label: Accession Number
  Group: ID
  Data: edsbas.7EDE2BF
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsbas&AN=edsbas.7EDE2BF
RecordInfo BibRecord:
  BibEntity:
    Languages:
      – Text: English
      – Text: Chinese
    Subjects:
      – SubjectFull: Video compression
        Type: general
      – SubjectFull: Digital video--Editing--Data processing
        Type: general
      – SubjectFull: Coding theory
        Type: general
      – SubjectFull: TK6680.5 .C479 2019eb
        Type: general
    Titles:
      – TitleFull: Effective methods in video coding and inpainting
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Chen, Zhe (author.)
      – PersonEntity:
          Name:
            NameFull: Cham, Wai-Kuen (thesis advisor.)
      – PersonEntity:
          Name:
            NameFull: Blu, Thierry (thesis advisor.)
      – PersonEntity:
          Name:
            NameFull: Chinese University of Hong Kong Graduate School. Division of Electronic Engineering. (degree granting institution.)
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 01
              Type: published
              Y: 2019
          Identifiers:
            – Type: issn-locals
              Value: edsbas
            – Type: issn-locals
              Value: edsbas.oa
ResultId 1