Multi-Scale Warping for Video Frame Interpolation
- Authors
- Choi, Whan; Koh, Yeong Jun; Kim, Chang-Su
- Issue Date
- 2021
- Publisher
- IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
- Keywords
- Interpolation; Kernel; Feature extraction; Convolution; Adaptive optics; Streaming media; Optical imaging; Video frame interpolation; convolutional neural network; multi-scale feature; kernel-based approach; deformable convolution; adaptive convolution
- Citation
- IEEE ACCESS, v.9, pp.150470 - 150479
- Indexed
- SCIE
SCOPUS
- Journal Title
- IEEE ACCESS
- Volume
- 9
- Start Page
- 150470
- End Page
- 150479
- URI
- https://scholar.korea.ac.kr/handle/2021.sw.korea/138658
- DOI
- 10.1109/ACCESS.2021.3126593
- ISSN
- 2169-3536
- Abstract
- A novel video interpolation network to improve the temporal resolutions of video sequences is proposed in this work. We develop a multi-scale warping module to interpolate intermediate frames robustly for both small and large motions. Specifically, the proposed multi-scale warping module deals with large motions between two consecutive frames using coarse-scale features, while estimating detailed local motions by exploring fine-scale features. To this end, it takes multi-scale features from the encoder and estimates kernel weights and offset vectors for each scale. Finally, it synthesizes multi-scale warping frames and combines them to obtain an intermediate frame. Extensive experimental results demonstrate that the proposed algorithm outperforms state-of-the-art video interpolation algorithms on various benchmark datasets.
- Files in This Item
- There are no files associated with this item.
- Appears in
Collections - College of Engineering > School of Electrical Engineering > 1. Journal Articles
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.