TimeLens: Event-based Video Frame Interpolation

TL;DR

TimeLens is a novel method that combines event-based and flow-based techniques to improve video frame interpolation, especially in highly dynamic scenes, achieving significant PSNR improvements over existing methods.

Contribution

We introduce TimeLens, a new approach that leverages both synthesis and flow information from event cameras for superior frame interpolation.

Findings

Up to 5.21 dB PSNR improvement over state-of-the-art methods.

Effective in highly dynamic and low-texture scenarios.

Provides a new large-scale dataset for dynamic scene interpolation.

Abstract

State-of-the-art frame interpolation methods generate intermediate frames by inferring object motions in the image from consecutive key-frames. In the absence of additional information, first-order approximations, i.e. optical flow, must be used, but this choice restricts the types of motions that can be modeled, leading to errors in highly dynamic scenarios. Event cameras are novel sensors that address this limitation by providing auxiliary visual information in the blind-time between frames. They asynchronously measure per-pixel brightness changes and do this with high temporal resolution and low latency. Event-based frame interpolation methods typically adopt a synthesis-based approach, where predicted frame residuals are directly applied to the key-frames. However, while these approaches can capture non-linear motions they suffer from ghosting and perform poorly in low-texture regions with few events. Thus, synthesis-based and flow-based approaches are complementary. In this work, we introduce Time Lens, a novel indicates equal contribution method that leverages the advantages of both. We extensively evaluate our method on three synthetic and two real benchmarks where we show an up to 5.21 dB improvement in terms of PSNR over state-of-the-art frame-based and event-based methods. Finally, we release a new large-scale dataset in highly dynamic scenarios, aimed at pushing the limits of existing methods.