How do AI systems turn pixels into useful information? Intro to Computer Vision AI, Part 5, introduces the concepts behind image understanding and shows how they connect to computer vision tools in Microsoft Azure and Microsoft Foundry. This course is Part 5 of 8 in the complete Microsoft AI Fundamentals AI-901 class. It builds on the earlier parts of the series and focuses on computer vision. After completing this part, continue to Part 6: Extraction AI, where you will explore how AI extracts useful information from content.
You will begin with the main computer vision tasks: image classification, object detection, semantic segmentation, and contextual image analysis. Learn how computers represent images as arrays of pixels and how filters help extract visual features. The lessons explain convolutional neural networks (CNNs) and vision transformers (ViTs), including the roles of feature maps, image patches, embeddings, and attention. Next, explore how multimodal models combine images and language to interpret visual content. You will examine image generation and the basic idea behind diffusion models, then connect these concepts to image analysis and generation in Microsoft Foundry. The course also introduces Azure vision capabilities such as image captions, visual tags, optical character recognition (OCR), face detection, and video analysis.
Business examples show how computer vision can support healthcare, manufacturing, and retail. Alongside these applications, you will consider responsible AI, including fairness, privacy, and the effects of bias in vision systems.
