Deep Learning/CNN Models
Resnet (2015)
Learning-Developer
2019. 11. 5. 11:55
1. Introduction
2. Realted Work
Residual Representations
Shortcut Connections
3. Deep Residual Learning
3.1 Residual learning
3.2 Identity Mapping by Shortcuts
3.3 Network Architectures
Plane Network
Residual Network
3.4 Implementation
4. Experiments
4.1 ImageNet Classification
Plain Network
Residual Network
Identity vs Projection Shortcuts
Deeper bottleneck Architectures
50-layer ResNet
101-layer and 152-layer ResNets
Comparisons with State of the art Methods
4.2 CIFAR-10 and Analysis
4.3 Object Detection on PASCAL and MS COCO
