30-Hrs Live Online Instructor-led Training on
APPLIED DEEP LEARNING & COMPUTER VISION ON IMAGE & VIDEO
Total Duration: 30 Hours |10-days | 3-Hours/-day
Training Highlights :
5+ Hands-on AI Projects 30+ Hours live online Instructor-led Training.
covered during the training.
Training includes: Certificate of completion by Eduxlabs in
Projects, Training PPT's & Recording. Association e-Cell IIT-Hyderabad
Training Projects :
Project 1: Face, Eye, Smile Detection using OpenCV
Project 1: Face Recognition using OpenCV & K-Nearest Neighbours Algorithm Classification
Project 2: Facial Expression / Emotion Detection using Deep Learning (CNN)
Project 3: Face Mask Detection and Classification using Deep Learning CNN and OpenCV for real time
recognition.
Project 4: Object Detection using Computer Vision and Deep Learning
Project 5: Traffic Sign Image Classification for Self-Driving Car
Project 6: Social Distancing Detector using Computer Vision and Deep Learning API Keras
Project 7: Brain MRI Medical Image Segmentation using U-Net
Project 8: Human pose estimation using Deep Learning in OpenCV
30-HRS COURSE CONTENT
DEEP LEARNING & COMPUTER VISION APPLICATION ON IMAGE and VIDEO
Module 1: Introduction to Machine Learning
• Introduction to Artificial Intelligence
• Introduction to Machine Learning
• Artificial Intelligence vs Machine Learning
• Benefits of Machine Learning w.r.t. traditional methodologies
• Deep Learning - Introduction
• How DL is different from all other Machine Learning methods
• Supervised and unsupervised learning
• Classification and Regression in Supervised Learning
• Clustering and Association in Unsupervised Learning
• Algorithms that are used in these categories
• Types of Data – Numerical and Categorical
• Types of Numerical Data – Discrete and Continuous
• Types of Categorical Data – Nominal, Ordinal and Dichotomous
Module 2: Python Fundamentals [CORE]
• Overview of Python- Starting with Python
• Python data types: Primitive
• Core built-in data structures – Lists, Tuples, Dictionaries, Sets
• String, String built-in methods
• Data manipulation tools (Operators, Functions, Packages, control structures, Loops, arrays etc.)
• Loops and Conditional statements
• Python UDFs – def keywords
• Reading and writing files
• Python Keywords – Lambda Function
Module 3: Data Analysis – Visualization Using Python
• Introduction exploratory data analysis
• Descriptive statistics, Frequency Tables and summarization
• Univariate Analysis (Distribution of data & Graphical Analysis)
• Bivariate Analysis (Cross Tabs, Distributions & Relationships, Graphical Analysis)
• Creating Graphs- Bar/pie/line chart/histogram/ boxplot/ scatter/ density etc)
• Exploratory Data Analysis
• Visualize distribution of Age variable with Seaborn
• Seaborn Kernel Density Estimation (KDE) plot
• Seaborn Catplot(), Countplot() function
Module 4: Working with Images and Video
Computer Vision - OPENCV
• Introduction Computer Vision
• Basics - Syntaxes
• Reading an Image
• Drawing Over an Image
• Mouse Handling
Image Transformation
• Image Transformation
• RGB to HSV, GRAYSCALE
• Edge Detection
• Smoothing Image
• Resizing Image
• Color Filter using OpenCV
• Color Detection
• Contour Detection
• Affine Transformation
• Homography Transformation
• Perspective Transformation
Video Transformation
• Starting VideoCam
• Reading from a Video
• Read, Write and Display Videos
• Implementation of color filter using video
• Understanding Haarcascades
• Cascade Classifier
• Video Gradients
Module 5: Face Detection Using HAARCASCADE
• HaarCascades
• Detect Multiscale
• Cascade Classifiers
• Face Detection using HaarCascades
• Face Detection using Deep Learning
• Person Detection
• Eye and Smile Detection using OpenCV & Haar Cascades
Module 6: KNN (K- Nearest Neighbours)
• K- Nearest Network – Introduction
• How does the KNN algorithm work?
• Methods of calculating distance between points
• Methods to choose the K
• GridSearchCV– Choosing Hyper parameters
Module 7: Logistic Regression
• Assumptions
• Probability and Odd Ratio
• Logical Interpretation used in Classification
• Hypothesis – Sigmoid Equation, Logistic Equation
• Odds Ratio and Interpretation
• Model Specification
• Prediction Probability
• Model Parameter Significance Evaluation
• Model Optimization of threshold value
• Confusion Matrix, Accuracy
• Practical: Handwritten Digit Classification
Module 8: Module FACE RECOGNISATION using OpenCV
• Development of Human Face Recognition
• Collecting Image Samples
• Collecting the faces of the test data sets with labels
• Reading training samples & creating training classifiers
• Face detection from sample
• OpenCV Face Recognizer
• KNN Face Recognition
DENSE LAYER – ANN
Module 9: Single Layered Perceptron
• Perceptron
• Various activation functions in neural networks – Unit Step, Sigmoid, ReLu, Softmax, and hyperbolic
functions
• Single Layer Perceptron
Module 10: Artificial Neural Networks
• Perceptron
• ANN & Working
• Single Layer Perceptron Model
• Multilayer Neural Network
• Feed Forward Neural Network
• Cost Function Formation
• Activation Function
• Cost Function Optimization
• Applying Gradient Descent Algorithm
• Stochastic Gradient Descent
• Backpropagation Algorithm & Mathematical Modelling
• Programming Flow for backpropagation algorithm
• Use Cases of ANN – OR Gate
• Programming Single Layer Neural Networks using Python
Module 11: Artificial Neural Network Classification
• Logistic Regression – Theory
• Artificial Neural Network
• Dense Layer Network
• Classification Problems
• Training the model
• Hypothesis, Parameters & Hyperparameters
• Loss Function
• Model Optimization – Optimizers
• Activation Function - Sigmoid, Softmax
TENSORFLOW & KERAS
Module 12: TensorFlow
• TensorFlow library for AI
• Keras – High Level TensorFlow API
• Getting started with TensorFlow
• Installation & Setting up TensorFlow
• TensorFlow Data Structures
• Tensor board – Visualization
CNN
Module 13: CNNs (Convolutional Neural Networks)
• Convolutional Neural Network
• Understanding the architecture and use-cases of CNN
• Pooling Layer – Max Pooling, Average Pooling
• How to visualize using CNN
• How to fine-tune a convolutional neural network
• What is transfer learning
• Kernel filter, Feature maps, and pooling
• Deploying convolutional neural networks in TensorFlow
IMAGE SEGMENTATION
Module 14: Mask R-CNN
• Image Segmentation
• Instance Segmentation
• Region of Interest (RoI)
• Mask R-CNN Basic Architecture
• R-CNN Framework
• RoI Align
• Loss Function
• FCN vs MLP
Module 15: U-NETS
• Understanding Semantic Segmentation
• What is U-Net
• U-Net Architecture
• Contraction & Expansion Path in U-Net Architecture
• Biomedical Image Segmentation using U-Net
PROJECTS:
Project 1: Face, Eye, Smile Detection using OpenCV
Project 1: Face Recognition using OpenCV & K-Nearest Neighbours Algorithm Classification
Project 2: Facial Expression / Emotion Detection using Deep Learning (CNN)
Project 3: Face Mask Detection and Classification using Deep Learning CNN and OpenCV for real time
recognition.
Project 4: Object Detection using Computer Vision and Deep Learning
Project 5: Traffic Sign Image Classification for Self-Driving Car
Project 6: Social Distancing Detector using Computer Vision and Deep Learning API Keras
Project 7: Brain MRI Medical Image Segmentation using U-Net
Project 8: Human pose estimation using Deep Learning in OpenCV
Training includes:
Training will be taken by one Industrial Expert with the experience of 8-9 years in the industry and has delivered
more then 1000+ sessions in India and abroad. The training hour is 3 hours/ each day. Mode of training is
Instructor-led live online
• 30 Hours Instructor-led live online Hands-on based learning with Projects.
• Interactive Query Session.
• Soft copy of Training material, PPT's , Projects code & Training Recording .
• FDP/ short Training certificate of completion will be provided to each Attendee.
Who can attend?
• Training is best suitable for Engineering college faculty, Research scholar, Student & Working IT
Professional.
EduxLabs Team
(Esoir Business Solutions Gurugram)
M: +91-7053133032 | 8851533667
Email [email protected]| www.eduxlabs.com
