Introduction to On-Device AI - Syllabus

• Introduction
  • Overview of on-device AI and deployment
  • Course objectives and key concepts
• Why On-Device
  • Benefits of on-device deployment: reduced latency, privacy, efficiency
  • Real-world applications of on-device AI
• Deploying Segmentation Models On-Device
  • Steps to deploy image segmentation models on edge devices
  • Code examples for deploying real-time models
• Preparing for On-Device Deployment
  • Model conversion for PyTorch and TensorFlow
  • Ensuring device compatibility and runtime requirements
• Quantizing Models
  • Introduction to model quantization techniques
  • Code examples for reducing model size and enhancing speed
• Device Integration
  • Integrating models with device hardware: CPU, GPU, NPU
  • Optimizing performance with device-specific compute units
• Conclusion
  • Summary of key concepts and deployment strategies
  • Future directions for on-device AI
• Appendix - Building the App
  • Steps for integrating the model into an Android app
• Appendix - Tips and Help
  • Additional resources and troubleshooting tips