AI Course Check

Computer vision courses for training models or using Azure

For a free computer vision course, take fast.ai's Practical Deep Learning for Coders, which teaches vision with PyTorch, fastai and Hugging Face.

Courses
9 courses
Free to take
5 courses

To add image features on Azure instead, Microsoft Learn's free paths include hands-on computer vision exercises.

Should you learn to train vision models or use a cloud service?

If you want to train and tune your own models, take fast.ai; if you want an app to read images through a cloud service, start with Microsoft Learn.

fast.ai's Practical Deep Learning for Coders is free and intermediate level, and fast.ai describes it as "9 lessons, and each lesson is around 90 minutes long". Vision sits alongside NLP, tabular data and deployment, so the course goes wider than images. As the name says, it is for people who code. The catch: fast.ai doesn't say if there is a certificate.

Which computer vision courses use Microsoft Azure?

Each of these is free and treats computer vision as one topic among several, next to speech, text and agents.

All three prepare for Exam AI-901, Azure AI Fundamentals, which replaced AI-900.

Which certification includes computer vision?

The main option is the Microsoft Certified: Azure AI Apps and Agents Developer Associate, an intermediate certification for AI engineers who build on Microsoft Foundry with Python. Its proctored 120-minute exam, AI-103, covers computer vision next to generative AI, agents, text analysis and information extraction. The fee depends on the country where you sit it. The older Azure AI Engineer Associate (AI-102) is retired.

Microsoft's course for it, Develop AI apps and agents on Azure, is listed by Microsoft at "4 days"; it runs with an instructor or as four free self-paced learning paths. One area it covers is insights from visual data.

Is there a longer program that includes computer vision?

The IBM AI Engineering Professional Certificate includes computer vision among 13 courses that run from machine learning and Spark pipelines to deep learning with Keras, PyTorch and TensorFlow, then NLP and LLM apps. It is intermediate, aimed at data scientists and engineers, and given as "4 months at 10 hours a week" on a paid Coursera subscription. Choose it if you want an IBM certificate across the whole field; for vision on its own, fast.ai is free.

More computer vision courses

Recommended order: courses mainly about Computer vision first, then open now, wide audience, beginner level and lower cost. The 7 courses shown above are not repeated here unless you filter or search. How we order courses.

2 courses

Filter and sort

Level
Cost
Format
Role
Provider