Unsupervised Learning of Dense Optical Flow, Depth and Egomotion from Sparse Event Data

TL;DR

This paper introduces a lightweight, unsupervised neural network pipeline that accurately estimates dense depth, optical flow, and egomotion from sparse event data, enabling real-time robotics applications.

Contribution

It presents the first monocular, self-supervised pipeline for dense depth and optical flow from sparse event data with a compact neural network architecture.

Findings

Achieves 250 FPS inference on a single GPU.

Outperforms previous deep learning methods on event data.

Performs well during day and night conditions.

Abstract

In this work we present a lightweight, unsupervised learning pipeline for \textit{dense} depth, optical flow and egomotion estimation from sparse event output of the Dynamic Vision Sensor (DVS). To tackle this low level vision task, we use a novel encoder-decoder neural network architecture - ECN. Our work is the first monocular pipeline that generates dense depth and optical flow from sparse event data only. The network works in self-supervised mode and has just 150k parameters. We evaluate our pipeline on the MVSEC self driving dataset and present results for depth, optical flow and and egomotion estimation. Due to the lightweight design, the inference part of the network runs at 250 FPS on a single GPU, making the pipeline ready for realtime robotics applications. Our experiments demonstrate significant improvements upon previous works that used deep learning on event data, as well as the ability of our pipeline to perform well during both day and night.