Completed
Evolution of Vision Architectures
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Computer Vision Architecture Evolution: ConvNets to Transformers - Lecture 21
Automatically move to the next video in the Classroom when playback concludes
- 1 Introduction
- 2 Evolution of Vision Architectures
- 3 Hierarchy of SWIN vs. CNNs
- 4 Modernizing ConvNets
- 5 Modernizing ResNet
- 6 Macro Design Changes
- 7 Changing stage compute ratio
- 8 Changing stem to "Patch-ify"
- 9 Depthwise Conv. vs Self-Attention
- 10 Improvements
- 11 Inverted Bottleneck
- 12 Larger Kernel Sizes
- 13 Micro Designs (mD)
- 14 Replace RELU with GELU
- 15 Fewer Activation functions
- 16 Fewer Normalization Layers
- 17 Substituting BN with LN
- 18 Visualization
- 19 mD4- Improvement
- 20 Separate Downsampling Layer
- 21 Final ConvNext block
- 22 Networks for Evaluation
- 23 Training Settings
- 24 Machine Performance Comparison
