Joint disparity and motion estimation using optical flow for multiview Distributed Video Coding

Matteo Salmistraro, Lars Lau Raket, Catarina Brites, Joao Ascenso, Søren Forchhammer

    Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review


    Distributed Video Coding (DVC) is a video coding paradigm where the source statistics are exploited at the decoder based on the availability of Side Information (SI). In a monoview video codec, the SI is generated by exploiting the temporal redundancy of the video, through motion estimation and compensation techniques. In a multiview scenario, the correlation between views can also be exploited to further enhance the overall Rate-Distortion (RD) performance. Thus, to generate SI in a multiview distributed coding scenario, a joint disparity and motion estimation technique is proposed, based on optical flow. The proposed SI generation algorithm allows for RD improvements up to 10% (Bjøntegaard) in bit-rate savings, when compared with block-based SI generation algorithms leveraging temporal and inter-view redundancies.
    Original languageEnglish
    Title of host publicationProceedings of the European Signal Processing Conference (EUSIPCO 2014)
    Publication date2014
    ISBN (Print)9780992862619
    Publication statusPublished - 2014
    Event22nd European Signal Processing Conference - Lisbon Congress Center, Lisbon, Portugal
    Duration: 1 Sept 20145 Sept 2014
    Conference number: 22


    Conference22nd European Signal Processing Conference
    LocationLisbon Congress Center
    Internet address
    SeriesProceedings of the European Signal Processing Conference


    • Signal Processing and Analysis
    • Decoding
    • Disparity Estimation
    • Distributed Video Coding
    • Estimation
    • Image motion analysis
    • Interpolation
    • Joints
    • Motion Estimation
    • Multiview Video
    • Optical Flow
    • Silicon
    • Video coding


    Dive into the research topics of 'Joint disparity and motion estimation using optical flow for multiview Distributed Video Coding'. Together they form a unique fingerprint.

    Cite this