Human Pose Estimation (HPE) is the tracing of joints in a human body (like elbow, knee) in images and videos. While this is a next to impossible task for rule based algorithms, Deep Learning based Computer Vision can save the day. The applications of HPE ranges from finding out the right postures for physical workouts and suggesting healthy postures in healthcare related applications to extracting poses of human or human like characters from videos to build movement related libraries for animation. While currently the animation industry works by recording movements of humans wearing sensors on all the joints of interest, which are later transferred to the animated character. This is both an expensive and laborious exercise, which can be eased by bringing in Deep Learning models, which extracts poses from images and videos and stores them as libraries that can be applied on an animated character. This approach has the added advantage of being able to generate motion libraries out of videos and images that are already existing.