Overview
Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Learn to build and deploy a production-ready computer vision application in minutes using Roboflow Rapid through a practical industrial use case of counting yellow-capped vials on a conveyor belt. Discover how to upload video content and instantly generate object detection models, then enhance them with intelligent workflows for tracking and counting objects. Master the implementation of ByteTracker and Line Counter configurations for stateful object tracking across specific boundaries, and explore batch processing capabilities for testing and production deployment. Follow along as the tutorial demonstrates the complete workflow from raw video input to a fully functional counting application, covering video upload and processing, object class definition and initial detection setup, real-time model testing, model deployment and workflow creation, tracking block configuration for line-crossing detection, batch processing setup and execution, and results analysis and optimization.
Syllabus
00:00 Introduction - Count specific vials with Roboflow Rapid?
01:26 Uploading and Processing the Video
02:26 Defining Object Classes and Initial Detection
04:44 Real-Time Model Testing on a Video
05:25 Deploying the Model and Creating Workflows
07:09 Adding Blocks to Track Objects and Count Line Crossing
10:08 Configuring and Running Batch Processing
11:55 Reviewing the Results and Conclusion
Taught by
Roboflow