For personal health care applications, unobtrusive sensors, such as reflective photoplethysmography (rPPG), capacitive electrocardiography (cECG) or ballistocardiography (BCG), are used with increasing frequency. While these sensors provide more comfort for the user, they exhibit a lower signal-to-noise ratio (SNR) and especially suffer from motion artifacts (MAs). Therefore, methods for reliable detection of MAs as well as their classification in the case of, for example, sleep analysis, are researched. In this paper, support vector machines (SVM) are investigated for the detection and classification of motion artifacts. Two methods for classification with respect to eight classes of movements are presented. First, a direct multi-class classification, and second, a multi-class classification after perfect detection. Waveform-related features are created and used for the training of the SVMs. An openly available dataset (UnoVis data set) which provides nine recordings of six channels of signals with annotations for motion is used. For the binary classification, an accuracy, sensitivity and specificity of 91%, 71%, 97% (test set) and 92%, 73% and 98% (validation set) are achieved respectively. For the direct multi-class classification, the SVM’ s performance is rather poor with mean accuracies, sensitivities and specificities of 77%, 21% and 93% (test set) and 78%, 28% and 93% (validation set) respectively. Similar results were achieved for perfect prediction.