Transformers In Computer Vision - English Version

What you'll learn
What are transformer networks?
State of the Art architectures for CV Apps like Image Classification, Semantic Segmentation, Object Detection and Video Processing
Practical application of SoTA architectures like ViT, DETR, SWIN in Huggingface vision transformers
Attention mechanisms as a general Deep Learning idea
Inductive Bias and the landscape of DL models in terms of modeling assumptions
Transformers application in NLP and Machine Translation
Transformers in Computer Vision
Different types of attention in Computer Vision
Requirements
Practical Machine Learning course
Practical Computer Vision course (ConvNets)
Introduction to NLP course
Description
Transformer Networks are the new trend in Deep Learning nowadays. Transformer models have taken the world of NLP by storm since 2017. Since then, they become the mainstream model in almost ALL NLP tasks. Transformers in CV are still lagging, however they started to take over since 2020. We will start by introducing attention and the transformer networks. Since transformers were first introduced in NLP, they are easier to be described with some NLP example first. From there, we will understand the pros and cons of this architecture. Also, we will discuss the importance of unsupervised or semi supervised pre-training for the transformer architectures, discussing Large Scale Language Models (LLM) in brief, like BERT and GPT.This will pave the way to introduce transformers in CV. Here we will try to extend the attention idea into the 2D spatial domain of the image. We will discuss how convolution can be generalized using self attention, within the encoder-decoder meta architecture. We will see how this generic architecture is almost the same in image as in text and NLP, which makes transformers a generic function approximator. We will discuss the channel and spatial attention, local vs. global attention among other topics.In the next three modules, we will discuss the specific networks that solve the big problems in CV: classification, object detection and segmentation. We will discuss Vision Transformer (ViT) from Google, Shifter Window Transformer (SWIN) from Microsoft, Detection Transformer (DETR) from Facebook research, Segmentation Transformer (SETR) and many others. Then we will discuss the application of Transformers in video processing, through Spatio-Temporal Transformers with application to Moving Object Detection, along with Multi-Task Learning setup.Finally, we will show how those pre-trained arcthiectures can be easily applied in practice using the famous Huggingface library using the Pipeline interface.
Overview
Section 1: Introduction
Lecture 1 Introduction
Section 2: Overview of Transformer Networks
Lecture 2 The Rise of Transformers
Lecture 3 Inductive Bias in Deep Neural Network Models
Lecture 4 Attention is a General DL idea
Lecture 5 Attention in NLP
Lecture 6 Attention is ALL you need
Lecture 7 Self Attention Mechanisms
Lecture 8 Self Attention Matrix Equations
Lecture 9 Multihead Attention
Lecture 10 Encoder-Decoder Attention
Lecture 11 Transformers Pros and Cons
Lecture 12 Unsupervised Pre-training
Section 3: Transformers in Computer Vision
Lecture 13 Module roadmap
Lecture 14 Encoder-Decoder Design Pattern
Lecture 15 Convolutional Encoders
Lecture 16 Self Attention vs. Convolution
Lecture 17 Spatial vs. Channel vs. Temporal Attention
Lecture 18 Generalization of self attention equations
Lecture 19 Local vs. Global Attention
Lecture 20 Pros and Cons of Attention in CV
Section 4: Transformers in Image Classification
Lecture 21 Transformers in image classification
Lecture 22 Vistion Transformers (ViT and DeiT)
Lecture 23 Shifted Window Transformers (SWIN)
Section 5: Transformers in Object Detection
Lecture 24 Transformers in Object detection
Lecture 25 Obejct Detection methods review
Lecture 26 Object Detection with ConvNet - YOLO
Lecture 27 DEtection TRansformers (DETR)
Lecture 28 DETR vs. YOLOv5 use case
Section 6: Transformers in Semantic Segmentation
Lecture 29 Module roadmap
Lecture 30 Image Segmentation using ConvNets
Lecture 31 Image Segmentation using Transformers
Section 7: Spatio-Temporal Transformers
Lecture 32 Spatio-Temporal Transformers - Moving Object Detection and Multi-trask Learning
Section 8: Huggingface Vision Transformers
Lecture 33 Module roadmap
Lecture 34 Huggingface Pipeline overview
Lecture 35 Huggingface vision transformers
Lecture 36 Huggingface Demo using Gradio
Section 9: Conclusion
Lecture 37 Course conclusion
Section 10: Material
Lecture 38 Slides
Intermediate to Advanced CV Engineers,Intermediate to Advanced CV Researchers


Homepage