# Multi-camera object sync

**URL:** <https://community.stereolabs.com/t/multi-camera-object-sync/532>\
**Category:** Software\
**Created:** [October 4, 2021, 1:39pm UTC](https://community.stereolabs.com/t/multi-camera-object-sync/532 "2021-10-04T13:39:07Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![cpene1](https://community.stereolabs.com/user_avatar/community.stereolabs.com/cpene1/32/144_2.png) [@cpene1](https://community.stereolabs.com/u/cpene1)\
**Post date:** [October 4, 2021, 1:39pm UTC](https://community.stereolabs.com/t/multi-camera-object-sync/532/1 "2021-10-04T13:39:07Z")

</div>

Hi there!

I want to detect the same object with 2 cameras (from the left-side and right-side of a road), and somehow I need to know it’s the exact same object. So in simpler terms, I need 2 cameras to set the same id for the same object. Do you perhaps have any tips on how I should do this?

At first, I was thinking maybe set both cameras as static and (somehow) set the same initial world transform for both cameras. This way I can identify objects based on their 3D location, which will be the same (or almost the same?) for both cameras.

But I don’t feel this is the optimal way to do it and I’m not even sure this works the way I am thinking. Perhaps there’s a better solution? Thanks in advance for your support! 🙂

---

<div class="post-metadata">

**Author:** ![cpene1](https://community.stereolabs.com/user_avatar/community.stereolabs.com/cpene1/32/144_2.png) [@cpene1](https://community.stereolabs.com/u/cpene1)\
**Post date:** [October 8, 2021, 9:35am UTC](https://community.stereolabs.com/t/multi-camera-object-sync/532/2 "2021-10-08T09:35:50Z")

</div>

Would something like this work?

> [@Fusion of 3D Stereo-camera SLAM with 3D LiDAR SLAM](https://community.stereolabs.com/t/fusion-of-3d-stereo-camera-slam-with-3d-lidar-slam/464/2):
>
> Hi @FL399 the best way to align the two sensors is by providing a “tag” in the real world with known coordinates (usually the origin), detecting it with both the sensors, and calculating the relative positions with respect to it. Then you can use coordinate transform formulas to calculate the transform from the ZED to the Lidar or vice-versa. For the ZED you can use for example an ArUco tag (see [this repository](https://github.com/stereolabs/zed-aruco)), for the lidar you must find a way to detect it using only 3D information… [this pa…](https://arxiv.org/pdf/1908.10349.pdf)