ABSTRACT
The current advancement of technologies is that we are entering into the world of artificial intelligence (AI). AI-enabled computer systems are capable to understand natural language, vision, speech etc. AI techniques make the machine understand the visual world, and even perform better than humans. The computer vision (CV) makes the machine capable of doing complex tasks. Deep learning algorithms and their hardware implementations help the machine to understand vision. Computer vision based on deep learning has become very popular, efficient, and accurate in understanding the images and videos. Convolution neural networks are the main part of deep learning algorithms. CV algorithms are used to perform these tasks: including (i) object recognition which means that whether a particular image is present or not; (ii) object detection finds the name of object and its location, and (iii) scene understanding will convert the image into segments.CV subtasks and various algorithms of theses subtasks are studied in this seminar report. Basics of deep neural network are also covered in this text.
