Fig. 6
From: Wrist-worn wearables based on force myography: on the significance of user anthropometry
(1) Accuracy of trained models in identifying the correct number of static gestures occurring during the dynamic tasks. (2) Proportion of correct overlap between identified gestures and actual arm/wrist orientation in wrist flexion/extension and forearm pronation/supination motions. (3) Proportion of correct overlap between identified gestures and corresponding actual gesture in cylindrical grip, key grip, and tripod grip during dynamic tasks (SVM support vector machine, LDA linear discriminant analysis, ELM extreme learning machine, NN neural network)