Deep Learning for Computer Vision

Basic and Advanced Computer Vision

Language: english

Note: 0/5 (0 notes) 1,267 students New course

Instructor(s): Shrirang Korde

Last update: 2022-08-12

What you’ll learn

Basic and Advanced Computer Vision
Artificial Neural Network
Keras Tools, Keras API Support
Image Processing, CNN

Requirements

Python

Description

Computer vision is an area of deep learning dedicated to interpreting and understanding images. It is used to help teach computers to “see” and to use visual information to perform visual tasks

Computer vision models are designed to translate visual data based on features and contextual information identified during training. This enables models to interpret images and apply those interpretations to predictive or decision making tasks.

Image processing involves modifying or enhancing images to produce a new result. It can include optimizing brightness or contrast, increasing resolution, blurring sensitive information, or cropping. The difference between image processing and computer vision is that the former doesn’t necessarily require the identification of content.

Deep Learning is part of a broader family of machine learning methods based on artificial neural networks.

Deep-learning architectures such as deep neural networks, recurrent neural networks, convolutional neural networks have been applied to fields including computer vision, speech recognition, natural language processing, machine translation, bioinformatics, drug design, medical image analysis, material inspection and board game programs, where they have produced good results

Artificial neural networks (ANNs) were inspired by information processing and distributed communication nodes in biological systems. ANNs have various differences from biological brains.

Keras is the most used deep learning framework. Keras follows best practices for reducing cognitive load: it offers APIs, it minimizes the number of user actions required for common use cases, and it provides clear & actionable error messages.

Following topics are covered as part of the course

Introduction to Deep Learning
Artificial Neural Networks (ANN)
- Activation functions
- Loss functions
- Gradient Descent
- Optimizer
Image Processing
- Convnets (CNN), hands-on with CNN
Gradients and Back Propagation – Mathematics
- Gradient Descent
- Mathematics
Image Processing / CV – Advanced
- Image Data Generator
- Image Data Generator – Data Augmentation
- VGG16 – Pretrained network
- VGG16 – with code improvements
Functional API
- Intro to Functional API
- Multi Input Multi Output Model
Image Segmentation
Pooling
- Max, Average, Global
ResNet Model
- Resnet overview
- Resnet concept model
- Resnet demo
Xception
- Depthwise Separable Convolution
- Xception overview
- Xception concept model
- Xception demo
- Visualize Convnet filters

Who this course is for

Python programmers, Machine Learning aspirants, Deep Learning Aspirants

Course content

Introduction
- Contents
- Deep Learning Introduction
- Keras Introduction
Artificial Neural Network (ANN)
- Backpropagation
- Optimizer and Activation Functions
- Demo- Activation Function
- Loss Functions
Keras – Quick Start
- Prerequisite, Environment (Dev)
- Getting Started with Keras
- Demo with Keras
Functional API
- Functional API and Demo
- MIMO Functional API with Demo
Computer Vision – CNN, Handling Images
- CNN – Convolutional Neural Network
- Demo – CNN (Part 1)
- Demo – CNN (Part 2)
Mathematics – Gradients, Back Propagation
- Gradients, Back Propagation (Part 1)
- Gradients, Back Propagation (Part 2)
DL for Computer Vision
- Keras Image/Data Processing Tools
- Data Augmentation
- VGG16, Pretrained network
- Demo – VGG16
- Improvements with Data Generation – VGG16
Advanced DL for Computer Vision
- Transposed Convolution
- Keras APIs for Transposed Convolution
- Image Segmentation, Demo
- ResNet Overview
- Pooling, ResNet Demo
- Depthwise Separable Convolution
- Xception Concept Overview
- Xception Model Demo
- Keras Xception support
- Visualize Convnet filters for Xception
- Filters Interpretation
- Model Fitment – Design Issues