Human Pose Estimation is a crucial step towards understanding people in images and videos, aiming to localize human joints (keypoints) in 2D or 3D coordinates. The problem has been challenging due to factors like strong articulations, small joints, occlusions, clothing, and lighting changes. Classical approaches, such as the pictorial structures framework and deformable part models, have limitations, including not depending on image data. Deep Learning-based approaches have significantly improved performance, with CNNs being a key building block. The literature review covers various papers, including DeepPose, Efficient Object Localization Using Convolutional Networks, Convolutional Pose Machines, Human Pose Estimation with Iterative Error Feedback, Stacked Hourglass Networks for Human Pose Estimation, and Simple Baselines for Human Pose Estimation and Tracking. These models have achieved state-of-the-art performance on various benchmarks, such as the COCO dataset, and have been widely adopted in applications like Action recognition, Animation, Gaming, and Basketball player analysis. The papers also discuss common evaluation metrics, including PCP, PCK, PDJ, and OKS-based mAP, which are essential for measuring the performance of human pose estimation models.