by Thuc Trinh Le, Andrés Andres Almansa, Yann Gousseau, Simon Masnou
Abstract:
We propose a fast and automatic inpainting technique for high-definition videos which works under many challenging conditions such as a moving camera, a dynamic background or a long lasting occlusion. Built upon the previous work by Newson et al. [1] which optimizes a global patch-based function, our method makes a significant improvement, especially in motion preservation, by incorporating the optical flow in several stages of the algorithm. Moreover, code parallelization and a modification in the process of patches pairwise matching yield a significant reduction of computation time. Experimental results on both classical and challenging datasets show that our algorithm outperforms other state-of-the-art approaches.
Reference:
Motion-consistent video inpainting (Thuc Trinh Le, Andrés Andres Almansa, Yann Gousseau, Simon Masnou), In (ICIP) International Conference on Image Processing, IEEE, 2017.
Bibtex Entry:
@inproceedings{Le2017-ICIP,
Abstract = {We propose a fast and automatic inpainting technique for high-definition videos which works under many challenging conditions such as a moving camera, a dynamic background or a long lasting occlusion. Built upon the previous work by Newson et al. [1] which optimizes a global patch-based function, our method makes a significant improvement, especially in motion preservation, by incorporating the optical flow in several stages of the algorithm. Moreover, code parallelization and a modification in the process of patches pairwise matching yield a significant reduction of computation time. Experimental results on both classical and challenging datasets show that our algorithm outperforms other state-of-the-art approaches.},
Author = {Le, Thuc Trinh and Almansa, Andr{\'{e}}s Andres and Gousseau, Yann and Masnou, Simon},
Booktitle = {(ICIP) International Conference on Image Processing},
Date-Modified = {2019-09-06 16:06:17 +0200},
Doi = {10.1109/ICIP.2017.8296651},
Isbn = {978-1-5090-2175-8},
Issn = {15224880},
Keywords = {Optical flow,Patches,Video editing,Video inpainting,Video restoration,o,patches,video restoration},
Month = {sep},
Pages = {2094--2098},
Publisher = {IEEE},
Title = {{Motion-consistent video inpainting}},
Url = {https://hal.archives-ouvertes.fr/hal-01492536},
Year = {2017},
Bdsk-Url-1 = {https://hal.archives-ouvertes.fr/hal-01492536},
Bdsk-Url-2 = {https://doi.org/10.1109/ICIP.2017.8296651}}