Features

Cars, Bikes, Pedestrians, and more

OpenDataCam can detect Cars, Trucks, Busses, Bikes, Motorcycles, Pedestrians and many more common objects out of the box! In many cases, allowing you to get started without the need to train your own model. But if you like to customize the model, you can do that too!

Image of different object types such as cars, trams and donuts

Fine grained Counters

OpenDataCam lets you set multiple counters, to give you fine grained insights into what's happening. Each counter give you the time, type and direction when it gets triggered. When an object crosses multiple counters (e.g. when a car takes a turn at an intersection), OpenDataCam lets you track that too!

All of those can be configured from an easy to use User Interface.

OpenDataCam UI with counter line visible.

Trajectory Analysis

OpenDataCam lets you also understand how objects move by looking at their trajectories.

This allows you for example to study how Pedestrians cross a public square or how cars are crossing lanes.

Real-time or pre-recorded

OpenDataCam supports running from various video sources such as USB-Cameras, IP-Cameras, RTSP-Streams, YouTube Live-Streams etc.

If you already have pre-recorded video captures, you can import those into OpenDataCam, and analyze the video this way.

Edge Devices and Data Centers

OpenDataCam runs on a large variety of platforms, from small edge devices like NVIDIA Jetson Nano to large data centers. All you need is a GPU and CUDA.

You own the data

OpenDataCam by default runs in stand alone mode, so you stay in full control of the edata you collect.

No data leaves the device. Everything is stored on a database running inside OpenDataCam and can be downloaded by you at any time, via the Web Interface or API.