Processing may take a few seconds...



Recursive block splitting in feature-driven decoder-side depth estimation


[ 1 ] Wydział Informatyki i Telekomunikacji, Politechnika Poznańska | [ 2 ] Instytut Telekomunikacji Multimedialnej, Wydział Informatyki i Telekomunikacji, Politechnika Poznańska | [ S ] student | [ P ] employee

Scientific discipline (Law 2.0)

[2.3] Information and communication technology

Year of publication


Published in

ETRI Journal

Journal year: 2022 | Journal volume: vol. 44 | Journal number: iss. 1

Article type

scientific article

Publication language


  • decoder-side depth estimation
  • immersive video

EN This paper presents a study on the use of encoder-derived features in decoder-side depth estimation. The scheme of multiview video encoding does notrequire the transmission of depth maps (which carry the geometry of a three-dimensional scene) as only a set of input views and their parameters are com-pressed and packed into the bitstream, with a set of features that could make iteasier to estimate geometry in the decoder. The paper proposes novel recursiveblock splitting for the feature extraction process and evaluates different scenar-ios of feature-driven decoder-side depth estimation, performed by assessingtheir influence on the bitrate of metadata, quality of the reconstructed video,and time of depth estimation. As efficient encoding of multiview sequencesbecame one of the main scopes of the video encoding community, the experi-mental results are based on the “geometry absent” profile from the incomingMPEG Immersive video standard. The results show that the quality of synthe-sized views using the proposed recursive block splitting outperforms that ofthe state-of-the-art approach.

Pages (from - to)

38 - 50




License type


Open Access Text Version

final published version

Points of MNiSW / journal


Impact Factor

1.347 [List 2020]

This website uses cookies to remember the authenticated session of the user. For more information, read about Cookies and Privacy Policy.