This video will give you an overview about the course.

In this video, you will see installation of Python, Virtualenv, Jupyter, and OpenCV.

• Look at Downloading Python

• Install OpenCV and Jupyter Notebook

• Activate Virtual Environment

This video will read image from files.

• Import OpenCV

• Set the path

• Read the image

This video will show you how to resize and flip the images.

• Resize the image using OpenCV’s resize function

• Flip the image on horizontal and vertical axis

• Display using matplotlib

In this video, you will see how to save the images to disk.

• Import OpenCV

• Read image from the file

• Save in the current directory

This video will show images using WINDOW_NORMAL and WINDOW_AUTOSIZE flags.

• Import OpenCV, and reading image from the file

• Differentiate between flags

• Display the images

In this video, you will see how to draw lines, rectangle, and circle and adding text to image.

• Import OpenCV and read image from the file

• Use OpenCV’s line, rectangle, circle, and puttext functions

This video will show you how to flip the images depending upon user input from a keyboard.

• Import OpenCV and read image from the file

• Handle user’s input using selection structures

In this video, you will see how to draw circles depending on user’s click from a mouse.

• Import OpenCV and read image from the file

• Explore OpenCV’s Mouse Events

• Create and attach mouse callback to window

This video will capture and show frame from a camera.

• Use Built-in webcam to capture a frame

• Display the size of the captured frame

• Display the frame on screen

In this video, you will play frame stream from video.

• Import OpenCV and set the video path

• Play a video using while loop

In this video, you will see creation and initialization of matrices, access to its elements, pixels, and also how we can work with part of a matrix.

• Create a matrix and fill it with 255 as a value

• Set some individual pixels' values to white

• Set the second channel of all pixels inside a certain region to 255

This video will explain you how to change the data type of matrix elements from uint8 to float32and perform arithmetic operations without worrying about clamping values.

• Convert our image to one with floating data type elements

• Scale the elements of our image by 2 and clip the values

• Scale the elements of our image back

This video will explain you how to save any matrix (not only with image content) of any type and shape with NumPy's data persistence.

• Create a matrix with random values

• Save our random matrix

• Load it from the file

In this video you will see getting access to individual channels, swapping them, and performing algebraic operations.

• Import all necessary modules

• Swap the red and blue channels

• Swap the channels back and scale them differently

This video will explain you how to change the color space of an image.

• Load an image

• Convert the image to grayscale

• Convert the image to HSV

In this video, you will see how to compute histograms.

• Load the image as grayscale

• Compute a histogram function

• Plot histogram and display it

This video will explain you how to get rid of noise or dramatically decrease it.

• Load an image, convert it to floating-point, and scale it down

• Create noise in the image

• Apply GaussianBlur to the noisy image

In this video, you will see how to create your own linear filter and apply it to images.

• Read the test image

• Create an 11x11 sharpening kernel

• Filter the image using the kernel

In this video you will learn how to construct a Gabor filter kernel (useful for detecting edges in images) and apply it to an image.

• Read the test image

• Apply a simple binary threshold

• Apply adaptive thresholding

This video will show you conversion of grayscale image into a binary image using different thresholding approaches.

• Read the test image

• Apply erosion and dilatation

• Visualize the results

In this video, you will learn how to apply basic morphological operations to binary images.

• Create a binary image with a circle mask

• Create a binary image with a rectangle mask

• Combine the circle and rectangle masks using a bitwise AND operator

This video will explain you how to binarize grayscale images using the Otsu algorithm resulting an image with only two colors; white and black.

• Read the test image

• Estimate the threshold using Otsu's method

• And lastly, visualize the results

This video will explain you how to find external and internal contours in a binary image and learn about findContours function.

• Load the test binary image

• Find the external and internal contours

• Visualize the results

In this video, you will see how to extract connected components from a binary image using OpenCV’s built-in function.

• Open an image and find the connected components in it

• Open another image, find its Otsu mask, & get connected components

• Filter out the components with small areas and create a color image

In this video, you will see fitting lines and circles into two-dimensional point sets using fitEllipse and fitLine functions.

• Draw and randomly generate parameters of the ellipse

• Draw out the ellipse and generated points on the image

• Draw the result, and display the image

This video will calculate image moments using a simple OpenCV built-in function.

• Draw a test image—a white ellipse

• Compute the moments and print their values

• Perform a simple test to check whether the computed moments make sense

In this video, you will check whether a point is within a contour or not using pointPolygonTest function.

• Find the contours of the image and display them

• Define a callback function to handle a user click on the image

• Show the image with our mouse click handler

This video will explain you how to compute distance between the maps and depict the results with gradient around the circle.

• Draw a test image—a black circle

• Compute the distance from every point to the circle

• Visualize the results

In this video, you will see segmentation of pixels in different clusters using k-means algorithm.

• Open an image and convert it to Lab color space

• Perform k-means clusterization

• Display the original and segmented images together

In this video, we’ll use OpenCV’s built-in functions for warping an image using affine and perspective transformations.

• Define two functions to implement the process of points selection

• Compute the affine transformation

• Compute the perspective transformation

In this video, we’ll take two images from a window and try stitching them into Panorama.

• Load the images we're going to combine into a panorama

• Create a Panorama Stitcher and pass our images to it

• Display the result

This video will leverage OpenCV to remove defects such as marks, ink and spots from a photo with image inpainting.

• Define a class that encapsulates mask creation

• Create its defect version and a mask

• Apply inpaint algorithms

In this video, we’ll be finding corners in an image using Harris and FAST from OpenCV.

• Load an image and find its corners with cv2.cornerHarris()

• Create a FAST detector and apply it to the image

• Display the results

This video will compute descriptors for image keypoints using SURF and SIFT’s alternative called ORB

• Load modules and open image

• Use SURF, BRIEF, and ORB for feature detection and description

• Display the results

In this video, we’ll separate an object from other parts of a scene. In other words, where we want to create masks for the foreground and background.

• Open an image

• Define the mouse callback function

• Call cv2.grabCut() to create an object mask

In this video, we’ll learn how to detect edges in images using the Canny algorithm.

• Load the test image

• Detect the edges using the Canny algorithm

• Visualize the results

In this video, we’ll learn how to apply the Hough transform for the detection of lines and circles.

• Detect lines using the probabilistic Hough transform

• Detect circles using the Hough transform

• Visualize the results

In this video, we’ll consider the methods for finding objects on the image which correspond to some of the template.

• Load the test image

• Use cv2.matchTemplate function

• Visualize the results

In this video, we're going to apply the Median Flow object tracker to track objects.

• Open a video file and read its frame

• Create the Median Flow tracker and initialize it

• Visualize the results

In this video, we’ll learn how to use the different tracking algorithms implemented in the OpenCV tracking contribution module.

• Create the main window and loop over the different trackers

• Open the test video file and select an object

• Track until the video ends

In this video, we’ll find out how to compute and display the dense optical flow in several ways, using OpenCV functionality.

• Define the function to display the optical flow

• compute the dense optical flow using Gunnar Farneback's algorithm

• Visualize the results

In this video, we’ll learn how to detect chessboard and circle grid patterns.

• Load the test image with a chessboard

• Detect the chessboard pattern

• Detect the circle grid pattern

In this video, we’ll learn how to detect pedestrians using a pre-trained SVM model with HOG features.

• Load the test image

• Create the HOG feature computer and detector

• Visualize the results

In this video, we'll learn how to train KNN- and SVM-based digit recognition models.

• Prepare the train and test data (features and labels)

• Create a KNN & SVM model

• Evaluate the KNN and SVM models

In this video, we’ll learn how to detect faces using OpenCV’s HAAR cascades.

• Load the test image

• Invokes a detector to find all of the faces in the image

• Visualize the results

In this video, we’ll understand how to detect QR codes and remove perspective distortions to get a canonical view of the codes.

• Load the test image

• Find the outer corners of the QR code

• Show the code corners and un-warp the code to get a canonical view

In this video, you’ll learn how to convert an image to a tensor.

• Read the image

• Turn HWC into CHW

• Add a new dimension to make it NCHW

This video loads model from different deep learning frameworks.

• Import the modules

• Load models using OpenCV’s built-in functions

• Display data-type of returned objects

This video will preprocess the images before inserting it to neural network.

• Load an image

• Explore OpenCV’s blobFromImage function

• Use setInput() and forward() functions on CNN model

This video will explain what ImageNet is and how to download images from ImageNet.

• Provide keyword for the dataset required

• Fetch a list of URLs of all the images

In this video, we’ll learn what annotation is and when it’s used.

• Introduction to data annotation

• Download and run LabelImg tool

• Annotate an image using LabelImg tool

This video will give you a basic idea of dataset augmentation and how it’s applied

• Why we need Dataset augmentation

• Operations like flip, rotate and translations

In this video, we’ll try to classify the space-shuttle video accurately.

• Import the modules

• Load the model

• Pass each from the video to the model

We’ll detect the vehicles in a video with high accuracy.

• Use readNetFromCaffe function

• Set a confidence threshold

• Detect and display the results

This video segments the regions in a picture.

• Use readNetFromCaffe function

• Compute image with per-pixel class labels

• Segment and display the results

This video gives you an intro to Open Model Zoo. There are pre-trained deep learning models and samples.

• Study OpenVINO Toolkit

• Introduce Open Model Zoo

This video will show you ONNX (Open Neural Network Exchange), which is an open source artificial intelligence ecosystem.

• Learn more about ONNX

G-API is used to declare image processing task in form of expression and then submit it for execution.

• Learn more about G-API

Developing an Age and Gender Recognition system using OpenCV.

• Use cv2.dnn.readNetFromCaffe to load a trained model

• Use cv2.resize() and np.transpose() to subtract source image

• Classify Gender and Age Group

This video develops a Face Detection & Emotion Recognition system.

• Use cv2.CascadeClassifier() to load classifier for haarcascade

• Use Keras emotion classifier for emotion detection

• Visualize the results

Human detection task involves both identification of the presence of the objects and identification of the rectangular boundary surrounding each object.

• Use readNetFromCaffe function

• Set a confidence threshold to detect Human

• Detect and display the results

In this video, we give a brief study of pre-trained models for vehicle detection and Roadside Objects identification.

• Look at the advanced application with OpenVINO

A brief introduction to the course, its contents and flow. You will get a gist of what this course has to offer you.

An overview of Machine Learning on visual data like images and videos, its applications and impact.

• Brief discussion on ML

• Real Life examples of ML on visual data

Installing all dependencies to carry out the contents of this course fluidly

• Install Python 3.6

• Create a virtual environment

• Download packages (opencv, scikit-learn, etc)

In order to perform operations on images, we first need to know how to load them into our program

• Use the imread() function to read images

• Break a video into smaller frames and reading them

Learn about different types of color spaces and convert from one format to the other. Along with that you will see how to perform binary thresholding on grayscale images.

• Change colorspaces

• Use binary thresholding

• Resize and save an image

OpenCV provides functions to draw objects and shapes on images. Let’s explore these amazing function now!

• Draw Rectangles

• Drawg Circles

• Write text on images

Machine learning can be either done in a supervised way or in an unsupervised way. In this video, we look at the supervised approach.

• Learn different categories of Learning algorithms

• Understand what is Supervised Learning is

• Explore few examples

We explore two famous supervised learning algorithms and see how they work internally.

• Learn how KNN works

• Learn how SVM works

To put the ML algorithms to test, we use the Quick, Draw Dataset! from Google. We write some code to visualize the dataset and then look at the steps to establish a ML pipeline.

• Download the dataset

• Visualize the dataset

• Implement step to establish an ML pipeline

Use the scikit-learn library to train a ML model on the Quick, Draw! Dataset, using both KNN and SVM

• Preprocess dataset

• Build the training and testing sets

• Train the model, using both KNN and SVM

Differences between the supervised learning approach and unsupervised learning is discussed in this video

• Get introduced to Unsupervised Learning

• Understand the differences between the two

In this video, we will discuss in detail what clustering is and how the K Means algorithm works to train on an unlabeled dataset.

• Explore clustering

• Learn how to use Euclidean distance to measure similarity

• Look into the implementation of the K Means algorithm

We now use the scikit-learn library to perform binary classification using the K Means Algorithm on the Quick, Draw! Datatset.

• Use K Means in scikit-learn

• Visualize the clustered dataset

• Classify test images

In order to perform Object Detection using Color, we will first need to understand what a Histogram is. Then, we can apply BackProjection to segment out the image based on color.

• Understand what a Histogram is

• Learn what Backprojection is

In this video, we detect an object in a live video feed based solely on its color. We get its coordinates and draw bounding boxes as well

• Capture histogram of the object color

• Perform Segmentation using Backprojection

• Get Contours and bounding rectangles

In this video we discuss what a Haar Cascade is, and how it works.

• Get an in-depth understanding of what a Haar Cascade is

• Understand how does it work

In this video, we are going to use a pre-trained Haar Cascades to detect Cat faces in images and videos

• Load the classifier

• Perform detection on images

• Draw bounding boxes

How has Deep Learning impacted the way we look at Computer Vision and the role of Convolutional Neural Networks in doing so.

• Learn what a CNN is

• Know the important Applications of CNNs in Computer Vision

In order to detect objects, we need a deep learning model which has already been trained and integrate it with OpenCV

• Import a pre-trained model which is stored locally

• Use the model to run detection on images

TensorFlow is a Deep Learning library by Google and it provides an API for a user to train their own custom object detector.

• Get introduced to the API

• Implement the steps to train a custom object detector

In order to train our object detector, we first need to train it with a lot of images of our object. In this video, we download these images and annotate them.

• Download Images

• Understand what annotating is

• Use LabelImg / RectLabel to annotate images

In order to train our model, we need to convert the dataset into a format which Tensorflow understands – TFRecords. Then we choose and download a base model for re-training.

• Generate TFRecords

• Choose and download Base Model

• Start Training

Now that the training is complete, in order to use our model, we need to export it into a usable format.

• Export the Inference Graph

• Test the Model