将这段文字改成论文中的、方案分析中的采取的技术方法的部分：上述两份程序都是采用计算机视觉和机器学习进行数据处理和分析的。具体而言它们采用了mediapipe框架这是一个专用于机器学习中姿势估计手势识别等模型构建的框架。这个框架被广泛地应用于计算机视觉领域尤其是机器人学、计算机辅助诊断、人机交互等领域。训练处理视频的程序需要使用一个已知的数据集进行监督学习以训练机器模型进行动作分类。在这个程序中K-

The two programs mentioned above utilize computer vision and machine learning for data processing and analysis. Specifically, they employ the mediapipe framework, which is specialized for model construction in machine learning, such as pose estimation and gesture recognition. This framework has been widely used in the field of computer vision, particularly in robotics, computer-aided diagnosis, and human-computer interaction.

The program used to train and process videos requires supervised learning using a known dataset to train the machine model for action classification. In this program, the K-Nearest Neighbor (KNN) algorithm is applied to the training dataset and compared with the test data for classification. KNN is a supervised machine learning algorithm that considers each data sample point as a point in an n-dimensional space. Samples that are closer in distance are more likely to belong to the same class. Therefore, the algorithm calculates the k nearest points (a constant) between the classified sample and the sample data, and then classifies the classified data according to the labels of these points. The labels can be generated in various ways, such as Euclidean distance, cosine distance, and Manhattan distance.

The program used to extract data employs the OpenCV library to process video streams and obtain keypoint information. OpenCV is an open-source library widely used for image and video processing in computer vision, providing many computer vision algorithms and tools. Specifically, OpenCV uses the mediapipe Pose model to detect key points in the human body and obtain accurate human motion data.

However, calculating the position of key points is just the first step in extracting human motion. Further analysis and processing are required. In this program, many processing and calculation techniques are used, such as calculating the distance between two key points, calculating the angle between them, and calculating the time interval of key actions. This enables the extraction of higher-dimensional information and the formation of a complete feature vector, which facilitates subsequent training and classification.

In summary, the technical methods of these two programs are based on machine learning, computer vision, and related knowledge, utilizing modern algorithms and frameworks to effectively solve many problems in human motion recognition, particularly in expressing and recognizing the details of human posture changes. The accuracy of the recognition has also been significantly improved. In future research, we believe that these technical methods can be applied to a wider range of fields, such as video, health, and entertainment, promoting the development of human intelligence and technology