Video to Events: Recycling Video Datasets for Event Cameras

TL;DR

This paper introduces a method to convert conventional video datasets into synthetic event data, enabling training of event-based vision models and improving their performance on real event camera tasks.

Contribution

It presents a novel approach to generate synthetic event data from existing videos, facilitating training without the need for large real event datasets.

Findings

Models trained on synthetic events generalize well to real event data.

Synthetic data improves performance on object recognition and semantic segmentation.

Fine-tuning on real data enhances state-of-the-art results.

Abstract

Event cameras are novel sensors that output brightness changes in the form of a stream of asynchronous "events" instead of intensity frames. They offer significant advantages with respect to conventional cameras: high dynamic range (HDR), high temporal resolution, and no motion blur. Recently, novel learning approaches operating on event data have achieved impressive results. Yet, these methods require a large amount of event data for training, which is hardly available due the novelty of event sensors in computer vision research. In this paper, we present a method that addresses these needs by converting any existing video dataset recorded with conventional cameras to synthetic event data. This unlocks the use of a virtually unlimited number of existing video datasets for training networks designed for real event data. We evaluate our method on two relevant vision tasks, i.e., object recognition and semantic segmentation, and show that models trained on synthetic events have several benefits: (i) they generalize well to real event data, even in scenarios where standard-camera images are blurry or overexposed, by inheriting the outstanding properties of event cameras; (ii) they can be used for fine-tuning on real data to improve over state-of-the-art for both classification and semantic segmentation.