Computer Vision

9.

Deep Learning for Computer Vision

9.1.

Neural Network Fundamentals

9.1.1.

Perceptron Model

9.1.1.1.

Linear Separability

9.1.1.2.

Perceptron Learning Algorithm

9.1.2.

Multi-Layer Perceptrons

9.1.2.1.

9.1.2.2.

Universal Approximation Theorem

9.1.3.

Activation Functions

9.1.3.1.

Sigmoid Function

9.1.3.2.

Hyperbolic Tangent

9.1.3.3.

ReLU and Variants

9.1.3.3.1.

9.1.3.3.2.

9.1.3.3.3.

9.1.4.

Backpropagation Algorithm

9.1.4.1.

Chain Rule Application

9.1.4.2.

Gradient Computation

9.1.4.3.

Weight Update Rules

9.1.5.

9.1.5.1.

Mean Squared Error

9.1.5.2.

Cross-Entropy Loss

9.1.5.3.

9.1.6.

Optimization Methods

9.1.6.1.

Gradient Descent Variants

9.1.6.2.

Stochastic Gradient Descent

9.1.6.3.

Mini-Batch Gradient Descent

9.1.6.4.

Momentum Methods

9.1.6.5.

Adaptive Learning Rate Methods

9.1.6.5.1.

9.1.6.5.2.

9.1.6.5.3.

9.2.

Convolutional Neural Networks

9.2.1.

CNN Architecture Components

9.2.1.1.

Convolutional Layers

9.2.1.1.1.

Convolution Operation

9.2.1.1.2.

Kernels and Filters

9.2.1.1.3.

Stride and Padding

9.2.1.1.4.

9.2.1.2.

9.2.1.2.1.

9.2.1.2.2.

Average Pooling

9.2.1.2.3.

9.2.1.2.4.

Adaptive Pooling

9.2.1.3.

Fully Connected Layers

9.2.1.3.1.

Flattening Operation

9.2.1.3.2.

Dense Connections

9.2.2.

Regularization Techniques

9.2.2.1.

9.2.2.1.1.

Training vs Inference

9.2.2.1.2.

Dropout Variants

9.2.2.2.

Batch Normalization

9.2.2.2.1.

Internal Covariate Shift

9.2.2.2.2.

Normalization Process

9.2.2.2.3.

Learnable Parameters

9.2.2.3.

Layer Normalization

9.2.2.4.

Group Normalization

9.2.3.

CNN Architectures

9.2.3.1.

9.2.3.1.1.

Historical Significance

9.2.3.1.2.

Architecture Details

9.2.3.2.

9.2.3.2.1.

Deep Architecture

9.2.3.2.2.

ReLU Activation

9.2.3.2.3.

9.2.3.3.

9.2.3.3.1.

Deep and Narrow Design

9.2.3.3.2.

Small Filter Sizes

9.2.3.4.

9.2.3.4.1.

Inception Modules

9.2.3.4.2.

Network in Network

9.2.3.4.3.

Auxiliary Classifiers

9.2.3.5.

9.2.3.5.1.

Residual Connections

9.2.3.5.2.

Skip Connections

9.2.3.5.3.

Deep Network Training

9.2.3.6.

9.2.3.6.1.

Dense Connections

9.2.3.6.2.

9.2.3.7.

9.2.3.7.1.

Depthwise Separable Convolutions

9.2.3.7.2.

Efficient Architecture

9.2.3.8.

9.2.3.8.1.

Compound Scaling

9.2.3.8.2.

Neural Architecture Search

9.3.

Training Deep Networks

9.3.1.

Data Preparation

9.3.1.1.

Data Augmentation

9.3.1.1.1.

Geometric Transformations

9.3.1.1.2.

Photometric Transformations

9.3.1.1.3.

Advanced Augmentation

9.3.1.2.

Data Normalization

9.3.1.3.

Dataset Splitting

9.3.2.

Training Strategies

9.3.2.1.

Transfer Learning

9.3.2.1.1.

Pretrained Models

9.3.2.1.2.

Feature Extraction

9.3.2.1.3.

9.3.2.2.

Progressive Training

9.3.2.3.

Curriculum Learning

9.3.3.

Hyperparameter Optimization

9.3.3.1.

Learning Rate Scheduling

9.3.3.2.

Batch Size Selection

9.3.3.3.

Architecture Search

9.3.4.

Regularization and Generalization

9.3.4.1.

9.3.4.2.

9.3.4.3.

Model Ensembling

9.3.5.

Training Monitoring

9.3.5.1.

9.3.5.2.

Validation Metrics

9.3.5.3.

Gradient Analysis

Previous

8. Classical Machine Learning for Vision

Go to top

Next

10. Core Vision Tasks with Deep Learning