Comparative evaluation of static gesture recognition techniques based on nearest neighbor, neural networks and support vector machines

• DOI: 10.1007/s13173-010-0009-z
• Publication Date: 2010-06-08T09:38:00Z
• Authors: Alexandre Savaris; Aldo von Wangenheim
• Keywords: Artificial neural network; Artificial intelligence; Support vector machine; Cognitive Neuroscience; Gesture Recognition; Generalization; 02 engineering and technology; Pattern recognition (psychology); Gene; Mathematical analysis; Biochemistry; Gesture recognition; Gesture; Machine learning; FOS: Mathematics; 0202 electrical engineering, electronic engineering, information engineering; Eye Tracking in Human-Computer Interaction; Complement (music); Data mining; Complementation; Life Sciences; Computer science; Tactile Perception and Cross-modal Plasticity; Human-Computer Interaction; Continuous Recognition; Chemistry; Gesture Recognition in Human-Computer Interaction; Phenotype; Head Gesture Recognition; Computer Science; Physical Sciences; Classifier (UML); k-nearest neighbors algorithm; Hand Gesture; Mathematics; Computer Science(all); Neuroscience

Abstract It is a common behavior for human beings to use gestures as a means of expression, as a complement to speaking, or as a self-contained communication mode. In the field of Human–Computer Interaction, this behavior can be adopted to build alternative interfaces, aiming to ease the relationship between the human element and the computational element. Currently, various gesture recognition techniques are described in the technical literature; however, the validation studies of these techniques are usually performed isolatedly, which complicates comparisons between them. To reduce this gap, this work presents a comparison between three well-established techniques for static gesture recognition, using Nearest Neighbor, Neural Networks, and Support Vector Machines as classifiers. These classifiers evaluate a common dataset, acquired from an instrumented glove, and generate results for precision and performance measurements. The results obtained show that the classifier implemented as a Support Vector Machine presented the best generalization, with the highest recognition rate. In terms of performance, all methods presented evaluation times fast enough to be used interactively. Finally, this work identifies and discusses a set of relevant criteria that must be observed for the training and evaluation steps, and its relation to the final results.