Master AI-Powered Video Analytics on Edge Devices
Duration: 3 Days | Level: Intermediate to Advanced
Delivery Mode: Hands-on Labs & Instructor-Led Sessions
Platform: ZTL_Base_AI_SDK on Rockchip NPU-based SBCs
Course Overview
This intensive 3-day training equips engineers and developers with end-to-end skills for deploying AI-powered video analytics on single-board computers (SBCs). Using ZTL_Base_AI_SDK and Rockchip NPU-accelerated hardware, participants will learn to:
- Develop and optimize YOLOv8/YOLOv11 models for edge deployment
- Convert, compile, and deploy RKNN models for real-time inference
- Build complete video analytics pipelines—from data labeling to multi-camera inference
Key Learning Outcomes
✔ AI Model Development: Train and fine-tune YOLO models for object detection
✔ Edge Deployment: Convert ONNX to RKNN format for NPU acceleration
✔ Real-World Applications: Implement people counting, anomaly detection, and more
✔ Optimization: Multi-process inference and NPU parallel computing techniques
✔ End-to-End Pipeline: Dataset creation → Model training → Edge deployment
Training Agenda
Day 1: Foundations & Environment Setup
- Introduction to ZTL_Base_AI_SDK
- Supported hardware/OS, directory structure, and best practices
- PC Environment Setup
- Anaconda, RKNN Toolkit2, and SDK installation
- RKNN Model Deployment
- ONNX-to-RKNN conversion workflow
Day 2: Model Training & Optimization
- YOLOv8/YOLOv11 Training
- Dataset labeling (X-AnyLabeling) and augmentation
- Training and exporting to ONNX/RKNN
- NPU Acceleration
- Multi-NPU parallel processing for high-throughput inference
Day 3: Deployment & Advanced Applications
- Edge Inference Techniques
- USB camera integration and real-time object detection
- System Integration
- RTMP streaming, cloud connectivity, and alarm triggers
- Capstone Project
- Deploy a custom video analytics solution on SBCs
Who Should Attend?
- Embedded AI Engineers developing edge vision systems
- IoT Developers working on smart surveillance/retail analytics
- Technical Teams evaluating Rockchip NPU platforms
Why Choose This Training?
✅ Hands-on Labs: Work directly with Rockchip NPU hardware
✅ Industry-Driven Content: Focused on deployable solutions
✅ Expert Instructors: Learn from ZTL SDK developer user
✅ Certificate of Completion
Prerequisites
- Basic Python programming
- Familiarity with Linux commands
- No prior AI experience required (fundamentals covered)
Hardware Inclusive
Each registered trainee is provided with A576 board which can be kept after the training.
