Let’s talk
Back
Learning Series 08

Computer Vision

Explore image understanding, object detection, visual intelligence, image generation, and where vision models earn their place.

06
Topics
10m
Reading time
Intermediate
Level
Overview

This series covers how machines interpret images: core tasks, data preparation, generation, evaluation and running vision at scale.

Read AI Fundamentals first, particularly neural networks. References include the landmark papers for each technique.

What You’ll Learn

Images as numbers

Images are grids of pixel values. Convolutional neural networks slide learned filters over the image to build feature maps, from edges to objects. Residual connections made very deep networks trainable, and vision transformers apply attention to image patches.

In simple terms

A vision model learns what to look for by seeing many labelled images.

Key concepts
Real-world example

ResNet won the 2015 ImageNet challenge using residual connections; the Vision Transformer later showed attention alone can match convolutional networks at scale.

References
He et al. (2015) — Deep Residual LearningDosovitskiy et al. (2020) — Vision Transformer
Continue learning
09

Natural Language Processing

Understand language models, text processing, language understanding, and NLP applications.

—

Understanding Retrieval-Augmented Generation

A practical guide to grounding a model in your own content.