Skip to Main content Skip to Navigation
Conference papers

Motion-consistent video inpainting

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 par-allelization and a modification in the process of patches pair-wise 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.
Complete list of metadata
Contributor : Andrés Almansa Connect in order to contact the contributor
Submitted on : Monday, March 20, 2017 - 10:56:17 AM
Last modification on : Monday, June 28, 2021 - 2:26:07 PM
Long-term archiving on: : Wednesday, June 21, 2017 - 12:46:01 PM


Files produced by the author(s)



Thuc Le, Andrés Almansa, Yann Gousseau, Simon Masnou. Motion-consistent video inpainting. ICIP 2017: IEEE International Conference on Image Processing, Sep 2017, Beijing, China. ⟨10.1109/ICIP.2017.8296651⟩. ⟨hal-01492536⟩



Record views


Files downloads