Deep Event-Based Object Detection in Autonomous Driving: A Survey
Jie Jiang, Bingquan Zhou, Tianjian Zhou, Yi Zhong · 2024
Reliable Object detection plays a pivotal role in the realm of autonomous driving, requiring the precise and swift identification of objects within rapidly changing environments. Conventional frame-based cameras encounter difficulties in managing latency and bandwidth, highlighting the urgency for novel approaches. Event cameras have risen to prominence for their potential in autonomous vehicles, offering advantages such as minimal latency, extensive dynamic range, and reduced power usage. Nonetheless, harnessing the asynchronous and sparse nature of event data poses a challenge, especially when it comes to sustaining low-latency operations and creating efficient, lightweight detection frameworks. This paper offers a comprehensive review of object detection with event data for autonomous driving, highlighting the superior benefits that event cameras can provide.