3 · Deep learning

9. Computer Vision & Convolutional Networks

Convolutions, the ImageNet moment, modern backbones and vision transformers.

9 min read · 3 MCQs

Why convolutions

A convolution slides a small learned kernel across an image, sharing weights across positions. That gives translation equivariance and cuts parameters by orders of magnitude compared with a dense layer over pixels.

The architectures that mattered

AlexNet (2012) proved GPU-trained deep nets on ImageNet; VGG showed depth with small kernels; ResNet introduced residual blocks and reached hundreds of layers; EfficientNet balanced depth, width and resolution.

Beyond classification

Detection (YOLO, Faster R-CNN) predicts boxes; segmentation (U-Net, Mask R-CNN) predicts per-pixel labels. Vision transformers now split an image into patches and apply attention, matching or beating CNNs when pretraining data is plentiful.

Chapter quiz

3 questions · pass mark 75%
  1. 1. Weight sharing in a convolution provides…

  2. 2. ResNet's key contribution was…

  3. 3. A vision transformer treats an image as…

Answer every question to submit. Progress for ai-09 is saved in this browser.