Fast event-based Harris corner detection exploiting the advantages of event-driven cameras

TLDR

Consistent feature point detection underpins many computer vision tasks, and event‑based cameras, which respond only to changes, naturally enhance edges and simplify corner detection. The paper proposes an event‑based corner detection method that leverages the high temporal resolution, compressed data, and low latency of asynchronous neuromorphic cameras. The method adapts the Harris corner detector to event streams, replacing frames with μs‑resolution asynchronous events, and is evaluated on a controlled pattern and a real scenario using a DVS on the neuromorphic iCub robot. The detector achieves sub‑2‑pixel error consistently across motion speeds and directions, attains a speed‑proportional detection rate that surpasses frame‑based techniques for substantial motion, and reduces computational cost.

Abstract

The detection of consistent feature points in an image is fundamental for various kinds of computer vision techniques, such as stereo matching, object recognition, target tracking and optical flow computation. This paper presents an event-based approach to the detection of corner points, which benefits from the high temporal resolution, compressed visual information and low latency provided by an asynchronous neuromorphic event-based camera. The proposed method adapts the commonly used Harris corner detector to the event-based data, in which frames are replaced by a stream of asynchronous events produced in response to local light changes at μs temporal resolution. Responding only to changes in its field of view, an event-based camera naturally enhances edges in the scene, simplifying the detection of corner features. We characterised and tested the method on both a controlled pattern and a real scenario, using the dynamic vision sensor (DVS) on the neuromorphic iCub robot. The method detects corners with a typical error distribution within 2 pixels. The error is constant for different motion velocities and directions, indicating a consistent detection across the scene and over time. We achieve a detection rate proportional to speed, higher than frame-based technique for a significant amount of motion in the scene, while also reducing the computational cost.