Multi-Scale Warping for Video Frame Interpolation

Citations

WEB OF SCIENCE

5
Citations

SCOPUS

8

초록

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.

키워드

InterpolationKernelFeature extractionConvolutionAdaptive opticsStreaming mediaOptical imagingVideo frame interpolationconvolutional neural networkmulti-scale featurekernel-based approachdeformable convolutionadaptive convolutionMOTION ESTIMATION
제목
Multi-Scale Warping for Video Frame Interpolation
저자
Choi, WhanKoh, Yeong JunKim, Chang-Su
DOI
10.1109/ACCESS.2021.3126593
발행일
2021
유형
Article
저널명
IEEE Access
9
페이지
150470 ~ 150479