Neural Events: Discrete Asynchronous Autoencoders for Event-Based Vision

2026-06-18

AI Summary

This work addresses the challenge of efficiently compressing the sparse yet high-volume event streams produced by event cameras while preserving fine-grained temporal dynamics. The authors propose a discrete asynchronous autoencoder framework that introduces, for the first time, a learnable discrete neural event representation. In this approach, raw event streams are relabeled into a reduced set of highly informative neural events, each encoding local spatiotemporal context via a discrete code and triggering output only when the code changes. This method achieves substantial event rate reduction—down to approximately 50% of the original—while simultaneously enhancing semantic content and maintaining temporal precision. Experimental results demonstrate that the proposed framework matches or exceeds state-of-the-art performance on object detection and classification tasks.

Abstract

Event cameras capture dynamic scenes with exceptional temporal fidelity by representing them as a continuous stream of microsecond resolution \textit{events}. Each individual event, however, only carries minimal semantic value, merely signaling a localized brightness change. To derive meaningful signals, downstream algorithms need to quickly integrate cues from a potentially massive torrent of low-information events. Current architectures, however, are easily overwhelmed, struggling to balance capturing fine-grained temporal dynamics and maintaining a manageable data throughput. This paper proposes a framework to re-tokenize event streams into a small set of highly informative \textit{neural events}, each representing a local spatio-temporal context window with a discrete learnable code. Every time this code flips, a neural event is triggered, yielding a highly compressed data stream. We demonstrate that, across object detection and classification, networks trained on neural events are on par or surpass the performance of state-of-the-art approaches while reducing the event rate by a factor of 2.0.

Problem

• event-based vision
• event cameras
• data throughput
• temporal dynamics
• semantic representation

Innovation

• neural events
• event-based vision
• discrete autoencoders
• asynchronous processing
• data compression