Bringing Events into Video Deblurring with Non-consecutively Blurry Frames

Recently, video deblurring has attracted considerable research attention, and several works suggest that events at high time rate can benefit deblurring. Existing video deblurring methods assume consecutively blurry frames, while neglecting the fact that sharp frames usually appear nearby blurry frame. In this paper, we develop a principled framework D²Nets for video deblurring to exploit non-consecutively blurry frames, and propose a flexible event fusion module (EFM) to bridge the gap between event-driven and video deblurring. In D²Nets, we propose to first detect nearest sharp frames (NSFs) using a bidirectional LSTM detector, and then perform deblurring guided by NSFs. Furthermore, the proposed EFM is flexible to be incorporated into D²Nets, in which events can be leveraged to notably boost the deblurring performance. EFM can also be easily incorporated into existing deblurring networks, making event-driven deblurring task benefit from state-of-the-art deblurring methods. On synthetic and real-world blurry datasets, our methods achieve better results than competing methods, and EFM not only benefits D²Nets but also significantly improves the competing deblurring networks.