Key Deep Learning Architectures for Visual Object Recognition


This series provides an overview of some of the prominent neural network architectures. Reading through this guide and all supplemental materials should help you develop an understanding of the modern neural network architectures and the main ideas behind them.

Before starting, you should have some familiarity with the basics of neural networks, backpropagation algorithm, and gradient descent. To learn these, I recommend two amazing courses:

  1. Practical Deep Learning For Coders” from fast.ai
  2. CS231n: Convolutional Neural Networks for Visual Recognition” from Stanford

List of architectures in this guide

Sorted in chronological order.

Coming soon:

  • GoogLeNet (2014)
  • Inception (2014)
  • VGG (2014)
  • InceptionV2, InceptionV3 (2015)
  • ResNet (2015)
  • InceptionV4 (2016)
  • DenseNet (2016)
  • Xception (2016)
  • MobileNet (2017)
  • NASNet (2017)
  • SE-ResNet (2017)
  • MobileNetV2 (2018)

Related Posts

Key Deep Learning Architectures - ZFNet

Detailed overview of the ZFNet architecture.

Deep Learning Vision for Non-Vision Tasks

Learn about how creative data processing allows using the full power of deep learning vision for non-vision tasks.

Key Deep Learning Architectures - LeNet-5

Detailed overview of the LeNet-5 architecture.

Key Deep Learning Architectures - AlexNet

Detailed overview of the AlexNet architecture.

Stochastic Weight Averaging — a New Way to Get State of the Art Results in Deep Learning

SWA is a method to ensemble models in the weight space, allowing to achieve better generalization.

Understanding Hinton’s Capsule Networks. Part 4. CapsNet Architecture.

Taking apart the CapsNet, layer by layer.

Understanding Hinton’s Capsule Networks. Part 3. Dynamic Routing Between Capsules.

Dynamic routing is a novel algorithm that allows to train CapsNets.

Understanding Hinton’s Capsule Networks. Part 2. How Capsules Work.

Getting grasp of internal workings of capsules.

Understanding Hinton’s Capsule Networks. Part 1. Intuition.

This post provides basic intuition behind CapsNets.

Computer Vision-Based Vehicle Detection

In this project, I use computer vision methods and traditional machine learning models to build a real-time vehicle detection system.