IVC_IEEE CGA2020JanFeb.pdf (3.32 MB)
Interactive Video Completion
journal contributionposted on 2021-08-17, 07:23 authored by M Okabe, K Noda, Y Dobashi, Kenichi Anjyo
We propose an interactive video completion method aiming for practical use in a digital production workplace. The results of earlier automatic solutions often require a considerable amount of manual modifications to make them usable in practice. To reduce such a laborious task, our method offers an efficient editing tool. Our iterative algorithm estimates the flow fields and colors in space-time holes in the video. As in earlier approaches, our algorithm uses an L1L1 data term to estimate flow fields. However, we employ a novel L2L2 data term to estimate temporally coherent color transitions. Our graphics processing unit implementation enables the user to interactively complete a video by drawing holes and immediately removes objects from the video. In addition, our method successfully interpolates sparse modifications initialized by the designer. According to our subjective evaluation, the videos completed with our method look significantly better than those with other state-of-the-art approaches.
Preferred citationOkabe, M., Noda, K., Dobashi, Y. & Anjyo, K. (2020). Interactive Video Completion. IEEE Computer Graphics and Applications, 40(1), 127-139. https://doi.org/10.1109/MCG.2019.2950176
Journal titleIEEE Computer Graphics and Applications
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Image color analysisEstimationOptimizationIterative algorithmsManualsTask analysisImage and video inpaintinginteractive techniquesubjective evaluationoptical flowimage-based renderingSoftware EngineeringArtificial Intelligence and Image ProcessingElectrical and Electronic Engineering not elsewhere classified