Features of ailia Tracker

In this page, we present the features that are provided by both the C and the C# APIs.

Basic usage

Available models

You can use yolox_s for object detections model. And, you can use BYTETrack as object tracking model.

Interface

Tracking results

You can access to tracked object's information:

Item	Content
id	Tracking ID
category	Category of tracked object
prob	Inference probabilities when executing object detection
x	Top left X position of the bounding box(0 is the right edge, 1 is the left edge)
x	Top left Y position of the bounding box(0 is the top edge, 1 is the bottom edge)
w	Width of the bounding box
h	Height of the bounding box

Specialized features

Restriction of the target category

It is possible to restrict the target object category to be tracked. AILIATrackerObject has attribute category to indicate the object category. By checking this attribute, you can restrict what object category to be tracked. Human category is 0, for instance, only human object informations can be used by filtering AILIATrackerObject which category is 0.

GPU usage

On Windows and Linux, it is possible to perform inference on the GPU with cuDNN for object detection phase. In order to use cuDNN, please install the CUDA Toolkit and cuDNN from the NVIDIA website:

Please install the CUDA Toolkit by following the installer instructions. For cuDNN, after downloading it (and uncompressing it) please adjust the environment variable PATH to reflect its location. You need to register as NVIDIA developper in order to download these libraries.