 This paper proposes a novel approach to recognizing human body gestures through deep learning techniques. Specifically, it uses a combination of convolutional neural networks, CNNs, and long short-term memory, LSTM, networks to extract and classify features from video sequences. The resulting system achieves impressive accuracy in recognizing various types of gestures, such as confidence, cooperation, confusion, and uncomfortable. This article was offered by Amna Mujahid, Muhammad Aslam, Muhammad Usman Ghani Khan, and others.