Making Every Event Count: Balancing Data Efficiency and Accuracy in Event Camera Subsampling

TL;DR

This paper evaluates hardware-friendly event camera subsampling methods, introducing a causal density-based approach that improves classification accuracy by balancing data efficiency and task performance.

Contribution

It systematically compares six subsampling methods and proposes a simple causal density-based method that enhances event video classification accuracy.

Findings

Density-based subsampling improves accuracy in sparse regimes.

Performance is sensitive to hyperparameters and event count variance.

Key factors influencing subsampling effectiveness are identified.

Abstract

Event cameras offer high temporal resolution and power efficiency, making them well-suited for edge AI applications. However, their high event rates present challenges for data transmission and processing. Subsampling methods provide a practical solution, but their effect on downstream visual tasks remains underexplored. In this work, we systematically evaluate six hardware-friendly subsampling methods using convolutional neural networks for event video classification on various benchmark datasets. We hypothesize that events from high-density regions carry more task-relevant information and are therefore better suited for subsampling. To test this, we introduce a simple causal density-based subsampling method, demonstrating improved classification accuracy in sparse regimes. Our analysis further highlights key factors affecting subsampling performance, including sensitivity to hyperparameters and failure cases in scenarios with large event count variance. These findings provide insights for utilization of hardware-efficient subsampling strategies that balance data efficiency and task accuracy. The code for this paper will be released at: https://github.com/hesamaraghi/event-camera-subsampling-methods.