Applied Computer Vision using Deep Learning

Introduction to Computer Vision

You will get familiar with the world of Computer Vision. We will also understand how this field has emerged in the past few years.

Horizons of Computer Vision

You will get an overview of different types of problems that can be solved using computer vision like Image Classification, localization and object detection, Face detection, segmentation, image generation including many others.

Understanding different Pre-trained models

We will explore different pre-trained models that are commonly used to solve image classification problems. We will implement all these models and compare the results on Emergency vs Non-Emergency vehicle classification problem

Image Localization and Object Detection

You will understand how to detect single objects as well as multiple objects in an image. We will also cover different architectures like RCNN, Fast RCNN, and State-of-Art Deep Learning models like YOLO, SSD, Retinanet along with other pre-trained models for detection.

Pose Detection

In this project, we will build a model that can localize human joints (elbows, wrists, etc) in images.

Face Detection

In this project, you will discover how to utilize State-of-Art Deep Learning techniques for Detecting human faces in images.

Image Segmentation

You will learn how to segment objects in an image. You will understand different approaches that can be used to solve segmentation tasks along with State-of-Art Deep Learning models like Mask-RCNN. In this module, we will also solve a lane segmentation problem. One of the prime applications of lane segmentation is in self driving cars.

Image Generation

Generating images is one of the most interesting applications of computer vision and has recently gained interest among researchers. Here we will cover different types of generative models like Pixel RNN, Pixel CNN, Variational Autoencoders (VAE) and then will see how to use GANs for generating realistic images.