Computer Vision

  1. Core Vision Tasks with Deep Learning
    1. Image Classification
      1. Problem Formulation
        1. Single-Label Classification
          1. Multi-Label Classification
            1. Hierarchical Classification
            2. Network Architectures
              1. Standard CNN Architectures
                1. Attention Mechanisms
                  1. Vision Transformers
                  2. Evaluation Metrics
                    1. Top-1 Accuracy
                      1. Top-k Accuracy
                        1. Mean Average Precision
                        2. Advanced Techniques
                          1. Few-Shot Learning
                            1. Zero-Shot Learning
                              1. Domain Adaptation
                            2. Object Detection
                              1. Problem Definition
                                1. Localization vs Detection
                                  1. Bounding Box Representation
                                    1. Multi-Object Detection
                                    2. Two-Stage Detectors
                                      1. R-CNN Family
                                        1. R-CNN
                                          1. Fast R-CNN
                                            1. Faster R-CNN
                                            2. Region Proposal Networks
                                              1. Feature Pyramid Networks
                                              2. Single-Stage Detectors
                                                1. YOLO Family
                                                  1. YOLOv1 through YOLOv8
                                                    1. Architecture Evolution
                                                    2. SSD
                                                      1. Multi-Scale Detection
                                                        1. Default Boxes
                                                        2. RetinaNet
                                                          1. Focal Loss
                                                            1. Feature Pyramid Networks
                                                          2. Detection Components
                                                            1. Anchor Generation
                                                              1. Bounding Box Regression
                                                                1. Non-Maximum Suppression
                                                                  1. Hard Negative Mining
                                                                  2. Evaluation Metrics
                                                                    1. Average Precision
                                                                      1. Mean Average Precision
                                                                        1. Intersection over Union
                                                                      2. Semantic Segmentation
                                                                        1. Problem Formulation
                                                                          1. Pixel-Level Classification
                                                                            1. Dense Prediction
                                                                            2. Network Architectures
                                                                              1. Fully Convolutional Networks
                                                                                1. Encoder-Decoder Structure
                                                                                  1. Skip Connections
                                                                                  2. U-Net
                                                                                    1. Biomedical Applications
                                                                                      1. Symmetric Architecture
                                                                                      2. DeepLab Family
                                                                                        1. Atrous Convolution
                                                                                          1. Spatial Pyramid Pooling
                                                                                            1. CRF Post-Processing
                                                                                            2. PSPNet
                                                                                              1. Pyramid Scene Parsing
                                                                                                1. Global Context
                                                                                              2. Advanced Techniques
                                                                                                1. Multi-Scale Processing
                                                                                                  1. Attention Mechanisms
                                                                                                    1. Boundary Refinement
                                                                                                    2. Evaluation Metrics
                                                                                                      1. Pixel Accuracy
                                                                                                        1. Mean IoU
                                                                                                          1. Frequency Weighted IoU
                                                                                                        2. Instance Segmentation
                                                                                                          1. Problem Definition
                                                                                                            1. Object Detection plus Segmentation
                                                                                                              1. Instance-Level Masks
                                                                                                              2. Mask R-CNN
                                                                                                                1. Architecture Overview
                                                                                                                  1. Mask Branch
                                                                                                                    1. RoIAlign Operation
                                                                                                                    2. Alternative Approaches
                                                                                                                      1. YOLACT
                                                                                                                        1. SOLOv2
                                                                                                                          1. Panoptic FPN
                                                                                                                          2. Panoptic Segmentation
                                                                                                                            1. Stuff and Things
                                                                                                                              1. Unified Framework
                                                                                                                              2. Evaluation Metrics
                                                                                                                                1. Average Precision for Masks
                                                                                                                                  1. Panoptic Quality
                                                                                                                                2. Image Generation
                                                                                                                                  1. Generative Models Overview
                                                                                                                                    1. Discriminative vs Generative
                                                                                                                                      1. Likelihood-Based Models
                                                                                                                                        1. Implicit Models
                                                                                                                                        2. Variational Autoencoders
                                                                                                                                          1. Encoder-Decoder Architecture
                                                                                                                                            1. Latent Space Representation
                                                                                                                                              1. Variational Inference
                                                                                                                                                1. Reparameterization Trick
                                                                                                                                                2. Generative Adversarial Networks
                                                                                                                                                  1. Generator Network
                                                                                                                                                    1. Discriminator Network
                                                                                                                                                      1. Adversarial Training
                                                                                                                                                        1. Training Instabilities
                                                                                                                                                          1. GAN Variants
                                                                                                                                                            1. DCGAN
                                                                                                                                                              1. StyleGAN
                                                                                                                                                                1. CycleGAN
                                                                                                                                                                  1. Pix2Pix
                                                                                                                                                                2. Diffusion Models
                                                                                                                                                                  1. Forward Diffusion Process
                                                                                                                                                                    1. Reverse Diffusion Process
                                                                                                                                                                      1. Denoising Diffusion Probabilistic Models
                                                                                                                                                                        1. Score-Based Models
                                                                                                                                                                        2. Applications
                                                                                                                                                                          1. Image Synthesis
                                                                                                                                                                            1. Image-to-Image Translation
                                                                                                                                                                              1. Super-Resolution
                                                                                                                                                                                1. Inpainting