Coding Activities by Bhargav Venkat

Real-Time Face Tracking and Counting People

Students will implement real-time face tracking and counting using their webcam. They will detect faces, count the number of people in the frame, and display the count dynamically.

Face detection with open CV

This activity will guide students through accessing their computer’s camera and performing real-time face detection using OpenCV’s pre-trained Haar Cascade classifier.

Interactive Color Filters on Static Image

Students will manipulate color channels of a static image in real-time by pressing specific keys to apply different color filters, including color tints and intensity adjustments. The image will be updated instantly with each key press.

Interactive Edge Detection & Filtering

Explore edge detection techniques (Sobel, Canny, Laplacian) and apply noise reduction filters (Gaussian, Median) interactively. Students will experiment with parameters and observe their effects on real images.

Image Annotation with Shapes and Measurements

This activity involves annotating an image using OpenCV by drawing rectangles and circles to highlight regions of interest, connecting them with a line, and visualizing the image height using bi-directional arrows. Text annotations are added for clarity, making the image informative and visually structured.

Rotating and Adjusting Brightness

Rotate an image by 45 degrees and adjust its brightness to see the effects of basic arithmetic operations on images.

Color Conversions and Cropping

Convert an image from BGR to RGB and grayscale, then crop a region of interest.

Save & grayscale with OpenCv

In this activity, students will load an image, convert it to grayscale, resize it to a standard size (224x224), display the processed image, and optionally save it based on key press.

Display & resize image with OpenCV

In this activity, students will learn how to load an image using OpenCV, display it in a resizable window, and adjust the window size. They will also explore the properties of an image such as dimensions and channels.

House Price Prediction using Keras

Create a neural network that can help you predict the house price using Keras. All the data is numerical, free of null values, so data preprocessing is not required.

Try Different Activation functions

Open the link to the page given below. Consider you have been given a dataset - Pima Indians onset of diabetes dataset. This has patient medical record data for Pima Indians and whether they had an onset of diabetes of five years. It is a binary class classification problem (onset of diabetes as one or not as 0). This dataset consists of 3 features/columns, including diabetes onset. Now, set the training to testing ratio to 80%. Now add two hidden layers with 4 and 6 neurons respectively and create a google document to enter values of testing loss for the following entries - 1) Sigmoid activation function 2) Tanh activation function

Forward Propagation

Have a look at this image given below. Now consider the following values and then calculate the value of the Output Layer using forward propagation. Consider that this is a Classification problem, and use the activation function accordingly. Here are the values mentioned - X1 = 20 X2 = 30 X3 = 12 W1 = W2 = 0.6 W3 = 0.3 W4 = 0.7 Submit the link of a google document that will consist of an image of all the calculations done for getting the final output.