Getting Started with Computer Vision
summary
Builds a rock/paper/scissors classifier by uploading labeled images (about 138 scissors, 126 rock, 136 paper, 113 none), trains a Custom Vision project (training ~9 minutes), exports an ONNX model and a TensorFlow.js bundle, and runs inference in-browser every 100 ms prediction interval inside a Vue front end calling a simple backend API. Uses Azure Custom Vision and Azure Machine Learning for experiments (one model hit ~92% validation accuracy); demo works live but is biased to the training images.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 9:43When to use AI
- 15:07Programming
- 21:07Azure Cognitive Services
- 26:22Custom Vision
- 42:32Custom AI
- 50:25Rock Paper Scissors
- 52:47Recap
related talks
16mHow to get started with Computer Vision AutoML in Azure
44mGetting started with Azure Machine Learning services
01h 02mAI on the Edge
01h 04mBuilding and Training your own Custom Image Recognition AI
23mBuild a look alike engine with machine learning and Angular
33mBuilding end-to-end computer vision solutions from deep learning models
24mDon't want to miss out on the AI wave?
26mLet's Build a Doppelganger Game!