EventSync: Joint Recovery of Temporal Offsets and Relative Orientations for Wide-Baseline Event Cameras
Wanli Xing, Shijie Lin, Guangze Zheng, Linhan Yang, Yanjun Du, Jia Pan
Abstract
Event-Based cameras offer significant advantages due to their high temporal resolution and low power consumption. However, when deploying multiple such cameras, a critical challenge emerges: each camera operates on an independent time system, resulting in temporal misalignment that severely degrades performance in multi-event camera applications. Traditional hardware-based synchronization methods face significant limitations in compatibility and are impractical for wide-baseline configurations. We introduce EventSync, a software-based algorithm that achieves millisecond-level synchronization by exploiting the motion of objects in the cameras’ shared field of view, while simultaneously estimating the relative orientation between cameras. Our approach eliminates the need for physical connections, making it particularly valuable for wide-baseline deployments. Through comprehensive evaluation in both simulated environments and real-world indoor/outdoor scenarios, we demonstrate robust synchronization accuracy and precise extrinsic calibration across varying camera configurations, significantly outperforming existing methods. Code: https://github.com/wlxing1901/event-sync
BibTeX
@inproceedings{iros2025_eventsyncjointre,
title = {EventSync: Joint Recovery of Temporal Offsets and Relative Orientations for Wide-Baseline Event Cameras},
author = {Wanli Xing and Shijie Lin and Guangze Zheng and Linhan Yang and Yanjun Du and Jia Pan},
booktitle = {IROS 2025},
year = {2025}
}