.png){kind=icon}
返回
Deep Iterative Frame Interpolation for Full-frame Video Stabilization
.png){kind=icon}
OA
AI
DOI：10.1145/3363550.png)
.png){kind=icon}
.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Video stabilization is a fundamental and important technique for higher quality videos. Prior works have extensively explored video stabilization, but most of them involve cropping of the frame boundaries and introduce moderate levels of distortion. We present a novel deep approach to video stabilization that can generate video frames without cropping and low distortion. The proposed framework utilizes frame interpolation techniques to generate in between frames, leading to reduced inter-frame jitter. Once applied in an iterative fashion, the stabilization effect becomes stronger. A major advantage is that our framework is end-to-end trainable in an unsupervised manner. In addition, our method is able to run in near real-time (15 fps). To the best of our knowledge, this is the first work to propose an unsupervised deep approach to full-frame video stabilization. We show the advantages of our method through quantitative and qualitative evaluations comparing to the state-of-the-art methods.
Keyword:
Video stabilization
frame interpolation
deep learning
unsupervised learning
near real-time
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
9.5
论文数:
4.7K
被引数:
3.6W
机构
暂无机构信息
引用论文
分享收藏Deep Online Video Stabilization With Multi-Grid Warping Transformation Learning基于多网格扭曲变换学习的深度在线视频稳像
PREAI
分享收藏Wireless Underground Communications in Sewer and Stormwater Overflow Monitoring: Radio Waves through Soil and Asphalt Medium.png)
InformationOAAI
InformationIF0
OAAI分享收藏
查看更多内容