The Death of CNNs: How Vision Transformers Rewrote Computer Vision in 3 Years (Part 1: The CNN Era)
Last Updated on February 12, 2026 by Editorial Team
Author(s): Ampatishan Sivalingam
Originally published on Towards AI.
From AlexNet’s 2012 revolution to ResNet’s dominance, and why it all became obsolete overnight
In 2012, a neural network called AlexNet won the ImageNet challenge by a margin so absurd that researchers initially thought there was a scoring error. It beat the second-place entry by 10.8 percentage points, a gap unprecedented in computer vision competitions. Convolutional Neural Networks weren’t new, but AlexNet proved they could learn features humans couldn’t even describe.
The article explores the dramatic evolution in computer vision, primarily focusing on the rise and decline of Convolutional Neural Networks (CNNs) and the emergence of Vision Transformers (ViTs). It discusses the revolutionary impact of AlexNet, followed by a decade of dominance by CNNs characterized by incremental improvements, and culminates with the introduction of ViTs that reshaped the landscape, showing how they outperformed traditional architectures by learning directly from data without relying heavily on pre-defined features.
Read the full blog for free on Medium.
Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.
Published via Towards AI
Towards AI Academy
We Build Enterprise-Grade AI. We'll Teach You to Master It Too.
15 engineers. 100,000+ students. Towards AI Academy teaches what actually survives production.
Start free — no commitment:
→ 6-Day Agentic AI Engineering Email Guide — one practical lesson per day
→ Agents Architecture Cheatsheet — 3 years of architecture decisions in 6 pages
Our courses:
→ AI Engineering Certification — 90+ lessons from project selection to deployed product. The most comprehensive practical LLM course out there.
→ Agent Engineering Course — Hands on with production agent architectures, memory, routing, and eval frameworks — built from real enterprise engagements.
→ AI for Work — Understand, evaluate, and apply AI for complex work tasks.
Note: Article content contains the views of the contributing authors and not Towards AI.
Related posts
Recent Posts
