VibES: Induced Vibration for Persistent Event-Based Sensing

2025-08-26

AI Summary

Event cameras fail to generate sustained event streams in static or low-motion scenes due to insufficient brightness changes, severely limiting their applicability in computer vision. To address this, we propose a lightweight vibration-induction mechanism: a rotating unbalanced mass applies controlled, periodic micro-vibrations to the camera, inducing pixel-level brightness perturbations in static scenes and thereby eliciting continuous event generation. Concurrently, we design a dedicated motion-compensation algorithm that models and removes systematic motion artifacts introduced by the vibration. Our approach requires no hardware modification and significantly improves spatiotemporal consistency and reliability of event data. Experiments demonstrate superior performance over conventional event sensing without motion induction across downstream tasks—including image reconstruction and edge detection—achieving an average PSNR gain of 3.2 dB and a 12.7% improvement in edge F1-score.

Technology Category

• Computer Vision: Motion & Tracking
• Intelligent Robots: Manipulation
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Security and Privacy: Tracking, profiling, and countermeasures against them

Abstract

Event cameras are a bio-inspired class of sensors that asynchronously measure per-pixel intensity changes. Under fixed illumination conditions in static or low-motion scenes, rigidly mounted event cameras are unable to generate any events, becoming unsuitable for most computer vision tasks. To address this limitation, recent work has investigated motion-induced event stimulation that often requires complex hardware or additional optical components. In contrast, we introduce a lightweight approach to sustain persistent event generation by employing a simple rotating unbalanced mass to induce periodic vibrational motion. This is combined with a motion-compensation pipeline that removes the injected motion and yields clean, motion-corrected events for downstream perception tasks. We demonstrate our approach with a hardware prototype and evaluate it on real-world captured datasets. Our method reliably recovers motion parameters and improves both image reconstruction and edge detection over event-based sensing without motion induction.

Problem

• Inducing persistent event generation in static scenes
• Removing injected motion for clean perception tasks
• Improving image reconstruction and edge detection

Innovation

• Uses rotating unbalanced mass for vibration
• Combines motion-compensation pipeline for clean events
• Improves image reconstruction and edge detection