Multi-camera object tracking in traffic monitoring

Object tracking and behavior analysis based on the 3D reconstruction of the scene by analyzing video streams from multiple cameras

Spectra is a video-based solution capable of identifying and reconstructing the trajectory of all the objects moving in the environment over the field of view of all the cameras. By doing this also allows the detection of long-term vehicle behaviors and making inferences during all their journey.

 

Spectra is a traffic analysis software developed by Sprinx.ai
Unbounded potential

Spectra can overcome real-world scenarios, extracting features of the same object from several camera views, even if orientation and lighting conditions can create a massive variation in its camera projected aspect.
Spectra also integrates spatial-temporal constraints to correctly track the objects in the environment as they pass through the network of cameras. The system can estimate the relative position of the road lanes and surfaces (in terms of the complete reconstruction of different places and orientation) between the other cameras to compute which camera an object shall appear when exiting from another one.

Spectra traffic statistics software by Sprinx.ai
In every scenario

Spectra has been designed to face different tasks: single-camera calibration, multi-camera calibration, multi-camera tracking, data fusion (data association), global inference and data visualization.
Spectra processes live video streams to perform all these steps and obtain every object’s trajectory in the scene through all the cameras in the system. This is done considering problems such as occlusions, missing detections, and view changes, leveraging the capability to follow the objects at the single-camera and system-wide levels.

Easy to understand data

Spectra can produce more helpful information by aggregating all raw data, resulting in more consistent, accurate, and precise output.
The visualization of the data will allow the operators to understand and identify patterns and behaviors.