Multi-view video acquisition is widely used for reconstruction and free-viewpoint rendering of dynamic scenes by directly resampling from the captured images. This paper addresses the problem of optimally resampling and representing multi-view video to obtain a compact representation without loss of the view-dependent dynamic surface appearance. Spatio-temporal optimisation of the multi-view resampling is introduced to extract a coherent multi-layer texture map video. This resampling is combined with a surface-based optical flow alignment between views to correct for errors in geometric reconstruction and camera calibration which result in blurring and ghosting artefacts. The multi-view alignment and optimised resampling results in a compact representation with minimal loss of information allowing high-quality free-viewpoint rendering. Evaluation is performed on multi-view datasets for dynamic sequences of cloth, faces and people. The representation achieves >90% compression without significant loss of visual quality.


Optimal Representation of Multi-View Video
Marco Volino, Dan Casas, John Collomosse and Adrian Hilton


        AUTHOR = "Volino, M. and Casas, D. and Collomosse, J.P. and Hilton, A.",
        TITLE = "Optimal Representation of Multiple View Video",
        BOOKTITLE = BMVC14,
        YEAR = "2014",


Data used in this work can be found in the CVSSP Data Repository.


This research was supported by the EU-FP7 project RE@CT, BBC/EPSRC iCase Studentship and EPSRC Visual Media Platform Grant EP/F02827X. The authors would also like to thank Martin Klaudiny for providing the face and cloth datasets used in the evaluation.