Field
Computer Vision
Getting machines to make sense of images.
Computer vision was the first field deep learning conquered decisively. In 2012 AlexNet won ImageNet by a margin that ended the argument, and the modern era started there.
The field is organised by how much you need to know about an image. Image classification says what's in it. Object detection says where. Segmentation labels every pixel, which is what you need when the exact shape matters and you can afford the annotation. OCR reads the text — solved for clean scans, genuinely hard for everything else.
That last one carries the field's most useful warning: OCR's "solved" reputation causes real project failures, because handwriting, phone photos, and layout-carrying documents are different problems wearing the same name.
Start with Image Classification — everything else builds on it.
14 concepts in this field
Image Classification
Getting an AI to look at an image and say what it is — the foundational task of computer vision.
Object Detection
Finding where objects are in an image and what they are — drawing a labelled box around each one.
Image Segmentation
Labelling every pixel rather than drawing a box — what you need when the exact shape matters.
OCR (Optical Character Recognition)
Turning pictures of text into text — solved for clean documents, still genuinely hard for everything else.
Vision Transformer
Cut an image into patches, treat them as words, run a transformer — which works, and only if you have enough data.
Face Recognition
Identifying a person from their face — technically solved, and the single clearest case of a system that works well on average and fails on specific people.
Object Tracking
Following the same object across video frames — where a 200-line algorithm from 2016 still beats most deep learning.
Optical Flow
Estimating the motion of every pixel between two frames — and there's a proof you fundamentally can't, from looking at any one part of the image.
Depth Estimation
Working out how far away things are from an image — and from a single photo, the absolute scale is mathematically unknowable.
Pose Estimation
Finding the joints of a body in an image — solved well enough to be boring, and the applications are mostly about watching people.
Image Captioning
Describing an image in words — declared solved on benchmarks a decade ago, and the benchmarks were measuring the wrong thing.
Video Understanding
Recognising what's happening in video — where models score well on shuffled frames, which tells you what they actually learned.
Neural Radiance Fields
Reconstructing a 3D scene from photos by training a network to be the scene — a beautiful idea, and it was replaced in three years.
CLIP
Training on images and their captions until both live in one space — the model that connected vision to language, and the reason typing a prompt gets you a picture.