Self-supervised Learning of Event-guided Video Frame Interpolation for Rolling Shutter Frames

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lu, Yunfan, Liang, Guoqiang, Shen, Yiran, Wang, Lin
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908389845499904
author Lu, Yunfan
Liang, Guoqiang
Shen, Yiran
Wang, Lin
author_facet Lu, Yunfan
Liang, Guoqiang
Shen, Yiran
Wang, Lin
contents Most consumer cameras use rolling shutter (RS) exposure, which often leads to distortions such as skew and jelly effects. These videos are further limited by bandwidth and frame rate constraints. In this paper, we explore the potential of event cameras, which offer high temporal resolution. We propose a framework to recover global shutter (GS) high-frame-rate videos without RS distortion by combining an RS camera and an event camera. Due to the lack of real-world datasets, our framework adopts a self-supervised strategy based on a displacement field, a dense 3D spatiotemporal representation of pixel motion during exposure. This enables mutual reconstruction between RS and GS frames and facilitates slow-motion recovery. We combine RS frames with the displacement field to generate GS frames, and integrate inverse mapping and RS frame warping for self-supervision. Experiments on four datasets show that our method removes distortion, reduces bandwidth usage by 94 percent, and achieves 16 ms per frame at 32x interpolation.
format Preprint
id arxiv_https___arxiv_org_abs_2306_15507
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Self-supervised Learning of Event-guided Video Frame Interpolation for Rolling Shutter Frames
Lu, Yunfan
Liang, Guoqiang
Shen, Yiran
Wang, Lin
Computer Vision and Pattern Recognition
Robotics
Most consumer cameras use rolling shutter (RS) exposure, which often leads to distortions such as skew and jelly effects. These videos are further limited by bandwidth and frame rate constraints. In this paper, we explore the potential of event cameras, which offer high temporal resolution. We propose a framework to recover global shutter (GS) high-frame-rate videos without RS distortion by combining an RS camera and an event camera. Due to the lack of real-world datasets, our framework adopts a self-supervised strategy based on a displacement field, a dense 3D spatiotemporal representation of pixel motion during exposure. This enables mutual reconstruction between RS and GS frames and facilitates slow-motion recovery. We combine RS frames with the displacement field to generate GS frames, and integrate inverse mapping and RS frame warping for self-supervision. Experiments on four datasets show that our method removes distortion, reduces bandwidth usage by 94 percent, and achieves 16 ms per frame at 32x interpolation.
title Self-supervised Learning of Event-guided Video Frame Interpolation for Rolling Shutter Frames
topic Computer Vision and Pattern Recognition
Robotics
url https://arxiv.org/abs/2306.15507