Pose estimation is a critical computer vision task that has gained significant attention due to its wide-ranging applications in fields like augmented reality, human–computer interaction, robotics, and health care. This project aims to develop an advanced system capable of accurately estimating human body poses and tracking hand movements in real time using state-of-the-art artificial intelligence and deep learning techniques. The pose estimation component of the system leverages Convolutional Neural Networks (CNNs) to identify critical body joints and their spatial relationships within an image or video stream. The system achieves high precision and robustness by employing complex neural network architectures and sophisticated pose representation methods, even in challenging and dynamic environments. The project adopts a multi-stage approach combining deep learning with motion analysis for hand tracking. Initially, a CNN-based model localizes the hands within the input frames. Our model demonstrated impressive performance in accurately estimating human or object poses in this pose estimation study. The results indicate a high Mean Average Precision (mAP) score, signifying the model's robustness and reliability.