Driving in Spikes: An Entropy-Guided Object Detector for Spike Cameras

2025-11-19

AI Summary

To address degraded object detection performance in autonomous driving caused by motion blur and extreme illumination conditions, this paper proposes EASD, an end-to-end spiking-stream detection framework. EASD employs a dual-branch architecture that jointly models the sparse, asynchronous, and high-temporal-resolution output of event cameras by integrating global semantic features with an entropy-guided selective attention mechanism. To support this work, we introduce DSEC-Spike—the first simulation-based spiking-object detection benchmark tailored to driving scenarios. Experiments demonstrate that EASD significantly improves detection accuracy and robustness under challenging driving conditions while preserving the microsecond-level latency and ultra-high dynamic range inherent to event-based sensing. Our framework establishes a scalable paradigm for real-time, event-camera-driven perception in autonomous systems.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Computer Vision: Motion & Tracking
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications

Abstract

Object detection in autonomous driving suffers from motion blur and saturation under fast motion and extreme lighting. Spike cameras, offer microsecond latency and ultra high dynamic range for object detection by using per pixel asynchronous integrate and fire. However, their sparse, discrete output cannot be processed by standard image-based detectors, posing a critical challenge for end to end spike stream detection. We propose EASD, an end to end spike camera detector with a dual branch design: a Temporal Based Texture plus Feature Fusion branch for global cross slice semantics, and an Entropy Selective Attention branch for object centric details. To close the data gap, we introduce DSEC Spike, the first driving oriented simulated spike detection benchmark.

Problem

• Addresses motion blur and saturation in autonomous driving object detection
• Overcomes sparse spike camera data incompatibility with standard detectors
• Introduces dual-branch architecture for global semantics and object details

Innovation

• End-to-end spike camera detector with dual branch
• Temporal texture fusion for global cross slice semantics
• Entropy selective attention for object centric details