An HMM-Based Gesture Recognition Method Trained on Few Samples

/in , , /by
An HMM-Based Gesture Recognition Method Trained on Few Samples

Godoy, Vinicius and Britto, Alceu S. and Koerich, Alessandro and Facon, Jacques and Oliveira, Luiz E.S.

Proceedings – International Conference on Tools with Artificial Intelligence, ICTAI 2014

Abstract : This paper addresses the problem of recognizing gestures which are captured using the Kinect sensor in a educational game devoted to the deaf community. Different strategies are evaluated to deal with the problem of having few samples for training. We have experimented a Leave One Out Training and Testing (LOOT) strategy and an HMM-based ensemble of classifiers. A dataset containing 181 videos of gestures related to nine signs commonly used in educational games is introduced, which is available for research purposes. The experimental results have shown that the proposed ensemble-based method is a promising strategy to deal with problems where few training samples are available.

You might also like
A comparison of self-organizing neural networks for fast clustering of radar pulses
Distribution Matching Approaches To Medical Image Segmentation
[Seminar] Towards Practical Few-Shot Query Sets: Transductive Minimum Description Length Inference
Source-free domain adaptation for image segmentation
Emotion Recognition with Spatial Attention and Temporal Softmax Pooling
Predicting the Gene Status and Survival Outcome of Lower Grade Glioma Patients with Multimodal MRI Features
A syntax directed level building algorithm for large vocabulary handwritten word recognition
Pixel level image fusion for medical imaging: An energy minimizing approach