InfoCoBuild

Deep Learning for Visual Computing

Deep Learning for Visual Computing. Instructor: Prof. Debdoot Sheet, Department of Electrical Engineering, IIT Kharagpur. Deep learning is a genre of machine learning algorithms that attempt to solve tasks by learning abstraction in data following a stratified description paradigm using nonlinear transformation architectures. When put in simple terms, say you want to make the machine recognize Mr. X standing in front of Mt. E on an image; this task is a stratified or hierarchical recognition task. At the base of the recognition pyramid would be features which can discriminate flats, lines, curves, sharp angles, color; higher up will be kernels which use this information to discriminate body parts, trees, natural scenery, clouds, etc.; higher up it will use this knowledge to recognize humans, animals, mountains, etc.; and higher up it will learn to recognize Mr. X and Mt. E and finally the apex lexical synthesizer module would say that Mr. X is standing in front of Mt. E. Deep learning is all about how you make machines synthesize this hierarchical logic and also learn these representative features and kernels all by itself. It has been used to solve problems like handwritten character recognition, object and product recognition and localization, image captioning, generating synthetic images to self driving cars. This course would provide you insights to theory and coding practice of deep learning for visual computing through curated exercises with Python and PyTorch on current developments. (from nptel.ac.in)

Lecture 02 - Feature Extraction for Visual Computing

Concepts covered in this lecture: Introductory concepts; Texture characterization - statistical vs. structural; Co-occurrence matrices; Orientation histograms; Local binary patterns (LBP); Texture from Fourier features; Wavelets.


Go to the Course Home or watch other lectures:

Lecture 01 - Introduction to Visual Computing
Lecture 02 - Feature Extraction for Visual Computing
Lecture 03 - Feature Extraction with Python
Lecture 04 - Neural Networks for Visual Computing
Lecture 05 - Classification with Perceptron Model
Lecture 06 - Introduction to Deep Learning with Neural Networks
Lecture 07 - Introduction to Deep Learning with Neural Networks (cont.)
Lecture 08 - Multilayer Perceptron and Deep Neural Networks
Lecture 09 - Multilayer Perceptron and Deep Neural Networks (cont.)
Lecture 10 - Classification with Multilayer Perceptron
Lecture 11 - Autoencoder for Representation Learning and MLP Initialization
Lecture 12 - MNIST Handwritten Digits Classification using Autoencoders
Lecture 13 - Fashion MNIST Classification using Autoencoders
Lecture 14 - ALL-IDB Classification using Autoencoders
Lecture 15 - Retinal Vessel Detection using Autoencoders
Lecture 16 - Stacked Autoencoders
Lecture 17 - MNIST and Fashion MNIST with Stacked Autoencoders
Lecture 18 - Sparse and Denoising Autoencoder
Lecture 19 - Sparse Autoencoders for MNIST Classification
Lecture 20 - Denoising Autoencoders for MNIST Classification
Lecture 21 - Cost Functions
Lecture 22 - Classification Cost Functions
Lecture 23 - Optimization Techniques and Learning Rules
Lecture 24 - Gradient Descent Learning Rule
Lecture 25 - SGD and ADAM Learning Rules
Lecture 26 - Convolutional Neural Network Building Blocks
Lecture 27 - Simple CNN Model: LeNet
Lecture 28 - LeNet Definition
Lecture 29 - Training a LeNet for MNIST Classification
Lecture 30 - Modifying a LeNet for CIFAR
Lecture 31 - Convolutional Autoencoder and Deep CNN
Lecture 32 - Convolutional Autoencoder for Representation Learning
Lecture 33 - AlexNet
Lecture 34 - VGGNet
Lecture 35 - Revisiting AlexNet and VGGNet for Computational Complexity
Lecture 36 - GoogLeNet - Going Very Deep with Convolutions
Lecture 37 - GoogLeNet
Lecture 38 - ResNet - Residual Connections within Very Deep Networks and DenseNet Densely Connected Networks
Lecture 39 - ResNet
Lecture 40 - DenseNet
Lecture 41 - Space and Computational Complexity in DNN
Lecture 42 - Assessing the Space and Computational Complexity of Very Deep CNNs
Lecture 43 - Domain Adaptation and Transfer Learning in Deep Neural Networks
Lecture 44 - Transfer Learning a GoogLeNet
Lecture 45 - Transfer Learning a ResNet
Lecture 46 - Activation Pooling for Object Localization
Lecture 47 - Regional Proposal Networks (rCNN and Faster rCNN)
Lecture 48 - GAP + rCNN
Lecture 49 - Semantic Segmentation with CNN
Lecture 50 - UNet and SegNet for Semantic Segmentation
Lecture 51 - Autoencoders and Latent Spaces
Lecture 52 - Principle of Generative Modeling
Lecture 53 - Adversarial Autoencoders
Lecture 54 - Adversarial Autoencoder for Synthetic Sample Generation
Lecture 55 - Adversarial Autoencoder for Classification
Lecture 56 - Understanding Video Analysis
Lecture 57 - Recurrent Neural Networks and Long Short - Term Memory
Lecture 58 - Spatio-Temporal Deep Learning for Video Analysis
Lecture 59 - Activity Recognition using 3D-CNN
Lecture 60 - Activity Recognition using CNN-LSTM