A3Deep Learning & Computer Vision for Edge AI

Teach machines to see, and run your vision models on cameras, robots, and IoT devices.

Who it's for

Engineers and developers with Python programming skills and basic machine learning knowledge.

You'll walk away with

A vision model deployed on an edge device or live camera stream

Build deep learning models for vision tasks and deploy them efficiently on cloud and edge devices.

Format: Lectures + hands-on labs with edge hardware
Assessment: Pre- and post-tests; hands-on labs; edge deployment demo
Tools & platforms: PyTorchYOLOOpenCVONNXTensorRT / OpenVINONVIDIA Jetson / Raspberry Pi

Programme Curriculum

Explain the fundamentals of neural networks, CNNs, and vision transformers
Train and fine-tune models for image classification, object detection, and segmentation
Apply transfer learning and data augmentation when data is limited
Optimize and deploy vision models to edge and IoT devices

Module 1
Deep Learning Fundamentals
- Neural Networks in PracticeNeural networks; backpropagation; PyTorch; training best practices
Module 2
Computer Vision Models
- ArchitecturesConvolutional neural networks (CNNs); vision transformers (ViT)
- Vision TasksClassification; object detection (YOLO); segmentation; transfer learning
Module 3
Data & Multimodal Vision
- Data for VisionLabeling; data augmentation; synthetic data
- Vision-Language ModelsMultimodal models (e.g., CLIP); open-vocabulary detection
Module 4
Edge Deployment
- Optimizing for the EdgeModel compression; ONNX; TensorRT / OpenVINO
- Edge & IoT IntegrationDeploying on Jetson / Raspberry Pi; camera pipelines; IoT integration; capstone demo