IoT and Edge AI with OpenVINO: face recognition, object detection, pose estimation, and semantic segmentation.
IoT Scholarship Foundation
Edge AI with Intel OpenVINO
Key Models
- Face recognition: OpenCV 4.1.1
- Image classification: Deep learning with high accuracy
- Object detection: MobileNet SSD (~5 FPS)
- Pose estimation: Very fast and accurate
- Coral TPU: Good performance with modifications
- Intel Movidius Stick 2: Good with OpenCV and Python
OpenVINO Pipeline
- Pre-trained models from Open Model Zoo
- Model Optimizer: TF/PyTorch/Caffe → IR format
- Inference Engine: Run optimized IR models
- Edge deployment: Input streams, MQTT, web serving
Model Types
- Classification (yes/no, 1000 classes, 20K ImageNet)
- Detection (bounding boxes + classification)
- Segmentation (semantic: all same class; instance: separate objects)
- Pose estimation
- Text recognition
- GANs
IoT #OpenVINO #EdgeAI #FarshidPirahansiah
Source: Pirahansiah, F. “IoT Scholarship: Edge AI with OpenVINO”. pirahansiah.com, updated 2026-09-19. Canonical URL https://pirahansiah.com/farshid/content/course-iot-scholarship/ · markdown source course-iot-scholarship.md · how to cite, quote and link this site.