Secrets of Event-Based Optical Flow

TL;DR

This paper introduces a novel method for estimating optical flow from event camera data using an extended Contrast Maximization framework, achieving top performance on benchmarks and revealing issues with existing ground truth data.

Contribution

It develops a principled approach to optical flow estimation from event data, addressing overfitting, occlusions, and convergence, and demonstrates state-of-the-art results without supervised training.

Findings

Ranks first among unsupervised methods on MVSEC benchmark

Performs competitively on DSEC benchmark

Reveals issues in existing ground truth flow data

Abstract

Event cameras respond to scene dynamics and offer advantages to estimate motion. Following recent image-based deep-learning achievements, optical flow estimation methods for event cameras have rushed to combine those image-based methods with event data. However, it requires several adaptations (data conversion, loss function, etc.) as they have very different properties. We develop a principled method to extend the Contrast Maximization framework to estimate optical flow from events alone. We investigate key elements: how to design the objective function to prevent overfitting, how to warp events to deal better with occlusions, and how to improve convergence with multi-scale raw events. With these key elements, our method ranks first among unsupervised methods on the MVSEC benchmark, and is competitive on the DSEC benchmark. Moreover, our method allows us to expose the issues of the ground truth flow in those benchmarks, and produces remarkable results when it is transferred to unsupervised learning settings. Our code is available at https://github.com/tub-rip/event_based_optical_flow