Abstract
Syncalize is a method for precise synchronization of multiple cameras and audio devices with standard consumer hardware. A specially crafted video file embeds timecodes in both the visual and audio streams, which can be automatically detected in recordings. Our system accurately estimates the exposure start and end times for each video frame, enabling the calculation of exposure time, frame rate, and synchronization offsets. Syncalize enables automated track synchronization during post-processing in multimedia production workflows. It is particularly valuable for quality assessment and enhancing timing accuracy in robotic multi-sensor setups, for measurement campaigns, and ego--exo configurations. We demonstrate the robustness and accuracy of Syncalize in various experiments showcasing a wide variety of use cases, such as stereo vision and synchronous multimedia content creation using mobile phones. Syncalize provides a flexible and open source method for synchronization tasks leading to accurate evaluations and results.
| Originalsprache | Englisch |
|---|---|
| Titel | Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops |
| Seiten | 3896-3905 |
| Auflage | 2025 |
| Publikationsstatus | Veröffentlicht - 19 Okt. 2025 |
Research Field
- Assistive and Autonomous Systems