Deep Learning has become a state-of-the-art solution to all machine intelligence based problems due to its high accuracy and efficiency. It helps in making real time decisions in applications like Advanced Driver Assistance Systems (ADAS), Robots, Autonomous Vehicles, Industrial Automation, Aerospace and Defense specific tasks. For accurate decisions and real time behavior, a massive amount of data needs to be processed. To process these data, the Deep learning Hardware acceleration is needed in which an application offloads a high computational task into specialized hardware for achieving high efficiency when compared to software implementation in CPU alone. Deep learning acceleration in hardware utilizes specialized computing devices designed to significantly speed up the training and inference of deep learning models. These accelerators are optimized for the high computational demands and parallelism inherent in deep neural networks, surpassing general-purpose CPUs in efficiency, speed, and power consumption.