Motion Segmentation and Egomotion Estimation from Event-Based Normal Flow

TL;DR

This paper presents a novel event-based normal flow framework for motion segmentation and egomotion estimation that leverages high-temporal-resolution data and geometric constraints, enabling accurate, real-time scene understanding without full optical flow.

Contribution

It introduces an optimization pipeline that performs event over-segmentation, residual analysis, and hierarchical clustering to improve motion segmentation and egomotion estimation from event data.

Findings

Achieves accurate segmentation and motion estimation on EVIMO2v2 dataset.

Performs well at object boundaries and in real-time scenarios.

Does not require full optical flow computation.

Abstract

This paper introduces a robust framework for motion segmentation and egomotion estimation using event-based normal flow, tailored specifically for neuromorphic vision sensors. In contrast to traditional methods that rely heavily on optical flow or explicit depth estimation, our approach exploits the sparse, high-temporal-resolution event data and incorporates geometric constraints between normal flow, scene structure, and inertial measurements. The proposed optimization-based pipeline iteratively performs event over-segmentation, isolates independently moving objects via residual analysis, and refines segmentations using hierarchical clustering informed by motion similarity and temporal consistency. Experimental results on the EVIMO2v2 dataset validate that our method achieves accurate segmentation and translational motion estimation without requiring full optical flow computation. This approach demonstrates significant advantages at object boundaries and offers considerable potential for scalable, real-time robotic and navigation applications.