Automated People flow tracking | with Computer Vision and AI

While detecting objects (in this case people) with Computer Vision, is nowadays common knowledge and easily accessible to anyone, tracking is still a complex task.

Tracking objects requires assigning a univocal ID to each of them (after they are detected), and keeping track of that ID until the object is on the screen.
If we fail to keep the IDs we assigned to the objects, the counting will be wrong.

I will talk in this video about the main steps to build successfully such an application and how to address the challenges step by step.

➤ Courses:

➤ Follow me on:

➤ For business inquiries:

#computervision #flowtracking #ai

[imenecharabi]

best notifiction :D thanks for the video so intresting, so clear and well explained.

[HideBuz]

What do you think would be the hardware requirements? How many camera feeds can I process with one cpu?

[ThongVo-mvfq]

Can the graph be generated dynamically while we are collecting data, or must it be drawn after a certain period of time?
In my opinion, this idea is the best way to visualize the state of an area over time.

[__________________________]

Tell us some use cases of this project.

[__s__a__b__i__r__]

Can I get a link to the video you're using?

[mobinmohammadi]

It is a great project and idea. Is it possible to access the codes?