Recognizing Static Signs from the Brazilian Sign Language: Comparing Large-Margin Decision Directed Acyclic Graphs, Voting Support Vector Machines and Artificial Neural Networks
This work addresses the challenge of automatic sign language recognition, which is incremental as it compares existing methods on a specific dataset.
The paper tackled the problem of recognizing static signs from the Brazilian Sign Language (LIBRAS) by comparing large-margin decision directed acyclic graphs, voting support vector machines, and artificial neural networks, reporting statistically significant results using Cohen's Kappa statistic.
In this paper, we explore and detail our experiments in a high-dimensionality, multi-class image classification problem often found in the automatic recognition of Sign Languages. Here, our efforts are directed towards comparing the characteristics, advantages and drawbacks of creating and training Support Vector Machines disposed in a Directed Acyclic Graph and Artificial Neural Networks to classify signs from the Brazilian Sign Language (LIBRAS). We explore how the different heuristics, hyperparameters and multi-class decision schemes affect the performance, efficiency and ease of use for each classifier. We provide hyperparameter surface maps capturing accuracy and efficiency, comparisons between DDAGs and 1-vs-1 SVMs, and effects of heuristics when training ANNs with Resilient Backpropagation. We report statistically significant results using Cohen's Kappa statistic for contingency tables.