Beyond Sentiment Analysis: Object Detection with ML NET
summary
Claims up to 90% defect-detection improvement and saving $2 billion in counterfeit bills introduce why computer vision matters; hands-on shows using ML.NET Model Builder + Azure ML for GPU training, tagging images with VoTT, exporting an ONNX model and consuming it in a UWP app via WinML. Demo trains in ≈10 minutes, CLI/AutoML generated C# code, a community CLI library has ~17,000 downloads, detection uses a 0.5 confidence threshold, and real-time detection works but has visible frame-rate lag from WinML binding.
AI-written summary · about
chapters (15)
- 0:00Start
- 2:03Intro
- 3:13Agenda
- 5:28Projection
- 9:34Microsoft Ai Stack
- 10:34Ml Net
- 17:58Ml Net Cli
- 18:48Sentiment Analyzer
- 26:24Model Builder
- 27:55Visual Object Tagging Tool
- 28:56Demo
- 35:18Machine Learning Workspace
- 35:46Object Detection Target File
- 37:13Azure Machine Learning Studio
- 37:41Object Detection Model
related talks
42mStep into Machine Learning .NET
45mML: ML.NET: deep dive into dev AI
41mAI + Machine Learning for .NET Developers
59mMachine Learning in .NET (ML.NET)
52mML.NET - Machine Learning made for .NET
46mMachine Learning for .NET Developers
58mExploring Machine Learning for .Net Developers
53mBuilding Intelligent .NET MAUI Apps with ML.NET