Multiple view wide-baseline trimap propagation for natural video matting
Multiple view wide-baseline trimap propagation for natural video matting
This paper presents a method to estimate alpha mattes for video sequences of the same foreground scene from wide-baseline views given sparse key-frame trimaps in a single view. A statistical inference framework is introduced for spatio-temporal propagation of high-confidence trimap labels between video sequences without a requirement for correspondence or camera calibration and motion estimation. Multiple view trimap propagation integrates appearance information between views and over time to achieve robust labelling in the presence of shadows, changes in appearance with view point and overlap between foreground and background appearance. Results demonstrate that trimaps are sufficiently accurate to allow high-quality video matting using existing single view natural image matting algorithms. Quantitative evaluation against ground-truth demonstrates that the approach achieves accurate matte estimation for camera views separated by up to 180°, with the same amount of manual interaction required for conventional single view video matting. © 2010 IEEE.
Camera calibration, Camera view, High quality, Key frames, Manual interaction, Matting, Multiple views, Multiple-view, Natural image matting, Quantitative evaluation, Spatio-temporal, Statistical inference, Trimap, Video matting, Video sequences, Wide-baseline, Cameras, Estimation, Video recording, Motion estimation
82-91
Sarim, M.
1242380f-3b23-491b-ab75-85daccc93f92
Hilton, Adrian
12782a55-4c4d-4dfb-a690-62505f6665db
Guillemaut, J.-Y.
fcaefbd8-8ba1-4c43-978c-0b5d432a3285
Kim, H.
2c7c135c-f00b-4409-acb2-85b3a9e8225f
Takai, T.
53894c03-6042-4bc8-86fb-692161fd22bb
2010
Sarim, M.
1242380f-3b23-491b-ab75-85daccc93f92
Hilton, Adrian
12782a55-4c4d-4dfb-a690-62505f6665db
Guillemaut, J.-Y.
fcaefbd8-8ba1-4c43-978c-0b5d432a3285
Kim, H.
2c7c135c-f00b-4409-acb2-85b3a9e8225f
Takai, T.
53894c03-6042-4bc8-86fb-692161fd22bb
Sarim, M., Hilton, Adrian, Guillemaut, J.-Y., Kim, H. and Takai, T.
(2010)
Multiple view wide-baseline trimap propagation for natural video matting.
7th European Conference on Visual Media Production, , London, United Kingdom.
17 - 18 Nov 2010.
.
(doi:10.1109/CVMP.2010.18).
Record type:
Conference or Workshop Item
(Paper)
Abstract
This paper presents a method to estimate alpha mattes for video sequences of the same foreground scene from wide-baseline views given sparse key-frame trimaps in a single view. A statistical inference framework is introduced for spatio-temporal propagation of high-confidence trimap labels between video sequences without a requirement for correspondence or camera calibration and motion estimation. Multiple view trimap propagation integrates appearance information between views and over time to achieve robust labelling in the presence of shadows, changes in appearance with view point and overlap between foreground and background appearance. Results demonstrate that trimaps are sufficiently accurate to allow high-quality video matting using existing single view natural image matting algorithms. Quantitative evaluation against ground-truth demonstrates that the approach achieves accurate matte estimation for camera views separated by up to 180°, with the same amount of manual interaction required for conventional single view video matting. © 2010 IEEE.
This record has no associated files available for download.
More information
Published date: 2010
Additional Information:
cited By 1
Venue - Dates:
7th European Conference on Visual Media Production, , London, United Kingdom, 2010-11-17 - 2010-11-18
Keywords:
Camera calibration, Camera view, High quality, Key frames, Manual interaction, Matting, Multiple views, Multiple-view, Natural image matting, Quantitative evaluation, Spatio-temporal, Statistical inference, Trimap, Video matting, Video sequences, Wide-baseline, Cameras, Estimation, Video recording, Motion estimation
Identifiers
Local EPrints ID: 440575
URI: http://eprints.soton.ac.uk/id/eprint/440575
PURE UUID: cd808ea0-5953-48cc-a189-2e9dbec1beed
Catalogue record
Date deposited: 07 May 2020 16:37
Last modified: 17 Mar 2024 04:01
Export record
Altmetrics
Contributors
Author:
M. Sarim
Author:
Adrian Hilton
Author:
J.-Y. Guillemaut
Author:
H. Kim
Author:
T. Takai
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics