odundar / computer-vision

A set of tutorials for computer vision application development using OpenCV, Intel OpenVino and inference engines. Aim is to understand developing computer vision applications at the edge with additional hardware support e.g. GPU, Intel(R) Movidius.

Computer Vision Workshop

This is a workshop to teach reader understand fundamentals of computer vision and improve existing solutions with Intel's HW and SDK's.

Following agenda will be followed:

Agenda:

HW Requirements

Intel NUC (6th gen or above)

Intel Movidius Stick or UP Ai Core

SW Requirements

Ubuntu 16.04

python 3. (opencv-python, numpy, matplotlib)

Introduction to OpenCV

1. Computer Vision Fundamentals with OpenCV OpenCV Overview

2. Run OpenCV Realtime Object Detection with DNN Object Detection

3. Realtime Object Detection with OpenCV RealTimeObjectDetection

Introduction to NCSDK

1. Introduction to Movidius, Myriad Movidius Overview

2. Introduction to NCSDK: get SDK, install, run samples. Movidius Overview

3. Realtime Object Detection with OpenCV using Movidius NCSDK Inferrence RealTimeObjectDetection

Introduction to OpenVino

OpenVino Presentations can be accessed from: https://github.com/intel-iot-devkit/smart-video-workshop/tree/master/presentations

1. Introduction to Intel OpenVino OpenVino Overview

2. OpenVino Model Optimizer Tutorial (Yolo Model with Caffe/Tensorflow) OpenVino Overview

3. RealTimeObjectDetection with OpenVino (Inference Engine, Media SDK, OpenCV)