Object Tracking Workflows for Video

​Learn how to count, trace, and monitor movement in your video streams using SORT and OC-SORT.

​Building reliable object tracking into your computer vision pipeline is essential for counting items, analyzing movement patterns, and monitoring video streams over time. However, maintaining consistent tracks when objects intersect, move unpredictably, or temporarily disappear behind obstacles can easily break traditional logic.

​In this live session, Roboflow Machine Learning Engineer Lee Clement will demonstrate how to solve these challenges using the newly integrated tracking blocks in Roboflow Workflows. Powered by the Trackers 2.3 package, these new tools allow you to drag and drop state-of-the-art tracking algorithms directly into your pipelines and then easily deploy them in the cloud or edge.

​You’ll also learn about the core differences between the available tracking algorithms, highlighting when to use SORT for high-speed, real-time pipelines and when to deploy OC-SORT to handle heavy occlusions and non-linear motion. We will then dive into a live demo, building a custom object tracking workflow from scratch to show you exactly how to configure these tools for your own video data.

​--

​Every week we dive into the world of computer vision and explore new tools, interesting projects, and tutorials. This webinar is open to everyone, we hope to see you there!

We'll also make time at the end for open Q&A.