Vision Class

Blog • Aparat • Telegram

Introduction to Machine learning and Deep learning for Computer vision, A course from Shahid Rajaie University (SRU) of Tehran, Held in winter and fall of 2018.

Course starts with an Introduction to Computer Vision with practical approach using opencv on python, then, continues with an Introduction to Learning Algorithms and Neural Networks. after that, Deep Neural Networks published after 2012 are studied and are implemented using python and Tensorflow, Keras, and FastAI Machine learning and deep learning frameworks.

Sessions

Tables consisting content of each semester in Persian can be found here: Winter 2018 • Fall 2018

Computer Vision

S1 - Course Introduction

🎯 Topics

Computer vision overview Course logistics

💡 Slides

Introduction PDF

📒 NoteBooks

1. Beginning

S2 - Opencv basics in python

🎯 Topics

Reading Images Color Spaces Displaying Images Saving Images

📒 NoteBooks

1. Reading, writing and displaying images
2. Grayscaling
3. Color Spaces 3-2. Extra

📝 Student notes

• Opencv Installation and startup
• Introduction to Opencv
• Introduction to Anaconda
• Git and getting last updates

S3 - Image manipulation (part 1)

🎯 Topics

Linear algebra Transform matrices Interpolation Methods

💡 Slides

Image manipulations (1) PDF • PPT

S4 - Image manipulation functions in OpenCV

🎯 Topics

Draw geometric shapes Transform matrices Translations Rotation Resizing Image pyramids Cropping

📒 NoteBooks

4. Drawing Images
5. Translations
6. Rotations
7. Scaling, resizing and interpolations
8. Image Pyramids
9. Cropping

📝 Student notes

• Image manipulations part 1
• Image manipulations part 2
• Image manipulations part 3

🎞 Videos

Aparat

S5 - Image manipulation (part 2)

🎯 Topics

Logical and Mathematical Operations in OpenCV Image masking in OpenCV Convolution and Correlation filters Moving Average Sharpening Filters in OpenCV

💡 Slides

Image manipulations (2) PDF • PPT

📒 NoteBooks

10. Arithmetic Operations
11. Bitwise Operations and Masking
12. Convolutions and Blurring
13. Sharpening

📝 Student notes

• Image manipulations part 4
• Image manipulations part 5
• Image manipulations part 6

🎞 Videos

Aparat

S6 - Binary Images, Thresholds, and Morphology Operation

🎯 Topics

Images Types Binary images, and Thresholds Thresholds in OpenCV Morphology (Dilation, Erosion, Opening, and Closing) Morphology in OpenCV

💡 Slides

Binary Images and Morphology PDF • PPT

📒 NoteBooks

14. Thresholding, Binarization & Adaptive Thresholding
15. Dilation, Erosion, Opening and Closing

📝 Student notes

• Binary images and Thresholding
• Morphology, opening and closing

🎞 Videos

Aparat

S7 - Edge Detection

🎯 Topics

Images Derivative, and Gradient Canny, and Sobel Edge Detections Edge Detection in OpenCv Perspective Transformation in OpenCv Affine Transforms Using Webcam in OpenCv

💡 Slides

Edge Detection PDF • PPT

📒 NoteBooks

16. Edge Detection & Image Gradients
17. Perspective & Affine Transforms
18. Using Webcam

📝 Student notes

• Edge detection part 1
• Edge detection part 2
• Affine Transforms in Opencv
• Using webcam in Opencv

🎞 Videos

Aparat

Machine Learning

S8 - Introduction to Machine Learning

🎯 Topics

What is ML Supervised Learning Unsupervised Learning Reinforcement Learning ML projects Steps Train-Test Split Model evaluation

💡 Slides

Introduction to Machine Learning PDF • PPT

📝 Student notes

• Introduction to Machine Learning
• Machine Learning workflow

S9 - Classifications, KNN, SciKit-learn • Introduction to Neural Network (1)

🎯 Topics

Perceptron Weights and Biases in Perceptron Activation Function Input Feature Array Multilayer Perceptron (MLP) Layers in MLP (input, hidden, and output)

💡 Slides

Simple Classifier (KNN) PDF • PPT

Introduction to Neural Networks PDF • PPT

📒 NoteBooks

19. Introduction to ML, and using Hoda Dataset
20. K Nearest Neighbor classification

📝 Student notes

• KNN classifier in scikit-learn
• Introduction to Neural networks (part 1)
• Introduction to Neural networks (part 2)

🎞 Videos

Aparat

S10 - Introduction to Neural Networks (2)

🎯 Topics

Loss Function (Coss Function) Gradient Descent, and Back Propagation Model Visualization

🎞 Videos

Aparat

🔗 links

Model Visualization and observing changes in number of each layer using Tensorflow Playground

S11 - Introduction to Implementing Recurrent Neural Networks using Keras

🎯 Topics

Recurrent, fully connected Networks in Keras Declaring Model Architecture Choosing Loss function, and Optimizer Model Evaluation on Test Set Predicting using Model

📒 NoteBooks

21. A Gentle Introduction to Keras – Simple neural network(MLP)

📝 Student notes

• Introduction to implementing sequential models in Keras

🎞 Videos

Aparat

Deep Learning

S12 - Introduction to Deep Neural Networks

🎯 Topics

Classification Tasks in Real-Life Invariant Object Recognition KNN, pros and cons Over-fitting Dropout Convolutional Neural Networks (CNN) CNNs vs. Classic methods ImageNet

💡 Slides

Introduction to Deep Learning & Convolutional Neural Networks PDF • PPT

📒 NoteBooks

22. Dropout

📝 Student notes

• Over-fitting and Dropout
• Introduction to Convolutional Neural Networks

🎞 Videos

Aparat

S13 - Introduction to Convolutional Neural Networks

🎯 Topics

Kernels: Convolutional Filters Learning kernels vs. Designing Fitlers Same and Valid Convolutions Paddings and strides Image Size before and after conv. 3D convolutions Multi-filter convolutions Convolutional Layers Parameters Pooling Layers LeNet

💡 Slides

Convolutional Neural Networks PDF • PPT

🎞 Videos

Aparat

S14 - Implementing Persian Handwritten Numbers Recognition using Keras

🎯 Topics

CNN Layers CNN pros and cons CNNs in Keras Conv2D and MaxPooling2D functions Flatten Method Models Summery

📒 NoteBooks

23. Convolutional Neural Network: Hoda + Keras

🎞 Videos

Aparat

S15 - Cat vs. Dog Binary Classification

🎯 Topics

Train-Test-Validation Split Data Generators in Keras Sigmoid and Softmax Step per Epoch Over-fitting

📒 NoteBooks

24. CNN cat vs. dog

📝 Student notes

• Training a Convolutional Neural Networks in Keras

🎞 Videos

Aparat

S16 - Review on Image Classification Architectures • Case Studies

🎯 Topics

Brain Architecture AlexNet VGGNet GoogLeNet ResNet

💡 Slides

Case Studies PDF • PPT

📝 Student notes

• Famous Convolutional Neural Networks Architectures

🎞 Videos

Aparat

📖 Reading Materials

AlexNet

VGGNet

GoogLeNet

ResNet

S17 - Data Augmentation

🎯 Topics

Preventing Over-fitting Data Augmentation in Keras

💡 Slides

Data Augmentation & Transfer Learning PDF • PPT

📒 NoteBooks

25. Data Augmentation

S18 - Transfer Learning (1)

🎯 Topics

Loading Pre-trained Models Transfer Learning in Keras

💡 Slides

Data Augmentation & Transfer Learning PDF • PPT

📒 NoteBooks

26. Loading Trained Model in Keras
27. Transfer LEarning - Feature Extraction

📝 Student notes

• Looking into Brains visual cortex
• Transfer Learning

S19 - Transfer Learning (2)

🎯 Topics

Implementing classification in keras conv. layers as Feature extraction Fine-tuning

📒 NoteBooks

27. Using a pretrained convnet
28. Transfer learning feature extraction
29. Transfer learning Fine tuning

S20 - Face Verification & Identification (1)

🎯 Topics

One-shot Learning Siamese Networks Triplet Loss

💡 Slides

Face PDF • PPT

S21 - Face Verification & Identification (2)

🎯 Topics

Center Loss A-softmax Loss

📒 NoteBooks

30. Face Recognition

📖 Reading Materials

A Discriminative Feature Learning Approach for Deep Face Recognition PDF SphereFace: Deep Hypersphere Embedding for Face Recognition PDF

S22 - Face Detection

🎯 Topics

Face Detection HAAR Cascade Wider Challenge MTCNN Face Detection Project Instructions

📒 NoteBooks

31. Face & Eye Detection
32. MTCNN Detection Sample Code

📖 Reading Materials

Joint Face Detection and Alignment using Multi-task Cascaded Convolutional Networks (MTCNN) PDF

🔗 links

Wider Face Challenge

S23 - Batch-Norm, Learning-Rate Decay, and Multi-Label Classification

🎯 Topics

Batch-Norm Learning-Rate Decay Multi-Label Classification in Keras

💡 Slides

Batch-Norm, Learning-Rate Decay, and Multi-Label Classification PDF • PPT

📒 NoteBooks

33. Keras Multi Label (part 1)

📝 Student notes

• Batch-Norm, Learning-Rate Decay
• Multi-Label Classification

🎞 Videos

Aparat

S24 - A Gentle Introduction to Tensorflow (1)

Multi-Label Classification (Continued) Tensorflow Low-level API Graphs, Constant Tensors, and Sessions in Tensorflow

💡 Slides

A Gentle Introduction to Tensorflow PDF • PPT

📒 NoteBooks

34. Keras Multi Label (part 2)

🎞 Videos

Aparat

S25 - A Gentle Introduction to Tensorflow (2)

🎯 Topics

Placeholders and Variables Feeding and Fetching Graphs

💡 Slides

Batch-Norm, Learning-Rate Decay, and Multi-Label Classification PDF • PPT

📒 NoteBooks

35. Intro to Tensorflow

🎞 Videos

Aparat

S26 - Implementing Dense and CNN networks for MNIST Dataset using Tensorflow

🎯 Topics

MNIST Dataset Fully-connected Layers CNN Layers

📒 NoteBooks

36. MNIST Dataset in Tensorflow
37. Fully Connected Network MNIST Tensorflow
38. CNN MNIST Tensorflow

🎞 Videos

Aparat

S27 - An even more Gentle Introduction to FastAI (1)

🎯 Topics

Finding Efficient Learning Rate Stochastic Gradient Descent with Restarts

💡 Slides

An even more Gentle Introduction to FastAI (1) PDF • PPT

📒 NoteBooks

39. Intro FastAi

🎞 Videos

Aparat

S28 - An even more Gentle Introduction to FastAI (2)

🎯 Topics

Global Pooling Adaptive Pooling Change Image Size Between Epochs

💡 Slides

An even more Gentle Introduction to FastAI (2) PDF • PPT

📒 NoteBooks

40. Breeds

🎞 Videos

Aparat

S29 - An even more Gentle Introduction to FastAI (3) and An Introduction to Recurrent Neural Networks

🎯 Topics

Multi-Label Classification in FastAI RNNs

💡 Slides

An Introduction to RNNs PDF • PPT

📒 NoteBooks

41. Planet Multi Label (part 3)

🎞 Videos

Aparat

S30 - An Introduction to Recurrent Neural Networks (2)

🎯 Topics

Forward Propagation Back Propagation Language Models LSTM Vanishing Gradient

💡 Slides

Recurrent Neural Networks PDF • PPT

📒 NoteBooks

42. Text Generation with Lstm

🎞 Videos

Aparat

S31 - RNNs, LSTM, and GRU

🎯 Topics

Vanishing Gradient LSTM Bidirectional RNNs GRU Deep RNNs Character Level Language Models in Keras

💡 Slides

RNNs, LSTM, and GRU PDF • PPT

📒 NoteBooks

43. Basic Text Classification

🎞 Videos

Aparat

S32 - Natural Language Processing

🎯 Topics

Word Embedding Analogy

💡 Slides

Recurrent Neural Networks PDF • PPT

🎞 Videos

Aparat

S33 - Word2Vec

🎯 Topics

Word2Vec Word Embedding Skip-grams Softmax Classification issues Negative Sampling

💡 Slides

Recurrent Neural Networks PDF • PPT

🎞 Videos

Aparat

S34 - GloVe

🎯 Topics

Glove Gender and Race Biases Using Embedding Vectors in Keras

💡 Slides

Recurrent Neural Networks PDF • PPT

📒 NoteBooks

44. Using Word Embeddings

🎞 Videos

Aparat

S35 - Word Analogy and Text Classification in Practice

🎯 Topics

Word Analogy Removing Biases Word Embedding Emoji Dataset

💡 Slides

Recurrent Neural Networks PDF • PPT

📒 NoteBooks

1. Analogy Using Embeddings
2. Debiasing Word Vectors
3. Text Classification - Emojify

🎞 Videos

Aparat

S36 - Text Generation

🎯 Topics

RNN Character Level Embedding Eager Execution in Tensorflow

📒 NoteBooks

48. Text Generation on Shahnameh Tensorflow

🎞 Videos

Aparat

S37 - Image Captioning

🎯 Topics

Image Captioning Keras

📒 NoteBooks

49. Image Captioning

🎞 Videos

Aparat

📦 Files

Required files for training model

S38 - Sequence to Sequence Models and Machine Translation

🎯 Topics

Seq2Seq Models Machine Translation

💡 Slides

Sequence to Sequence Models PDF • PPT

🎞 Videos

Aparat

S39 - Attention

🎯 Topics

NLP Machine Translation Attention Layer Keras

💡 Slides

Attention and Memory PDF • PPT

📒 NoteBooks

50. Machine Translation with Attention

40 - Speech Recognition and Trigger Word Detection

🎯 Topics

Spectrogram Attention CTC Trigger Word Detection RNNs

💡 Slides

Speech Recognition and Trigger Word Detection using RNNs PDF • PPT

📒 NoteBooks

51. [Trigger Word Detection](51-Trigger-word-detection/Trigger word detection - v1.ipynb)

🎞 Videos

Aparat

41 - Music Generation

🎯 Topics

Trigger Word Detection Collaborative Filtering Recommendation systems RNNs

📒 NoteBooks

52. Music Generation with LSTM

📦 Files

Excel used for Collaborative Filtering

42 - Style Transfer and Book Recommendation System

🎯 Topics

Recommendation Systems GANs

💡 Slides

Neural Style Transfer PDF • PPT

📒 NoteBooks

53. Recommendation System

Practices

• P1 - Basics
• P2 - Binary images and Morphology
• P3 - Edge detection, perspective transform, and using Webcam
• P4 - KNN classifier, and Intro to Neural Networks in datacamp
• P5 - Introduction to Neural Networks
• P6 - Implement and train classifiers, and using Google Colab
• P7 - Objects Recognition using webcam
• P8 - Intro to Tensorflow
• P9 - LSTM, GRU, and Word Embedding

quizzes

• Convolutional Neural Networks Basics
• Word Embeddings, Glove and Global pooling

Guests

• Dr. Reza Ebrahimpour on Business and Graduate Studies
• Mohammad Chanarin Nakhaie on FastAI Introduction
• Mohammad Ghodoosi on Spiky Neural Networks
• Abolfazl Mehdizadeh on Image Captioning

Teamworks

• Collecting Iranian Celebrities pictures Dataset (step 1)