Courses/Computer Vision

Computer Vision

CSE471

Prof. Makarand Tapaswi + Prof. Charu Sharma•Spring 2025-26•4 credits

Revision Notes/Unit 4 — Convolutional Neural Networks (CNNs)

Unit 4 — Convolutional Neural Networks (CNNs)

The CNN era — and the first 'real' CV topic of the course. Convolutional layers (params, RF, padding, stride, dilation); 1×1 convs; pooling; batch normalisation in CNNs; equivariance vs invariance. Architectures in chronological order: LeNet → AlexNet → VGG → Inception → ResNet → DenseNet → SENet → MobileNet → EfficientNet. Closes with backprop through convs/pools and CNNs for video / audio.

CNN Concepts + Architectures — LeNet to EfficientNet, BN, 1×1, Dilation, Receptive Field

22 min

A convolution layer is a stack of learned filters slid across an image — a CNN is a hierarchy of these, with each layer composing simpler features (edges → texture → parts → objects). The reason CNNs work in vision: WEIGHT SHARING (one filter, all spatial locations) gives **translation equivariance** (a shifted input produces a shifted output) and slashes parameters by ~1000× vs a dense network. The decade of CNN architecture research (2012–2020) was about how to go deeper, wider, or more efficient without breaking trainability.