Abstract
Our research focuses on the question of classifiers that are capable of processing images rapidly and accurately without having to rely on a large-scale dataset, thus presenting a robust classification framework for both facial expression recognition (FER) and object recognition. The framework is based on support vector machines (SVMs) and employs three key approaches to enhance its robustness. First, it uses the perturbed subspace method (PSM) to extend the range of sample space for task sample training, which is an effective way to improve the robustness of a training system. Second, the framework adopts Speeded Up Robust Features (SURF) as features, which is more suitable for dealing with real-time situations. Third, it introduces region attributes to evaluate and revise the classification results based on SVMs. In this way, the classifying ability of SVMs can be improved.Combining these approaches, the proposed method has the following beneficial contributions. First, the efficiency of SVMs can be improved. Experiments show that the proposed approach is capable of reducing the number of samples effectively, resulting in an obvious reduction in training time. Second, the recognition accuracy is comparable to that of state-of-the-art algorithms. Third, its versatility is excellent, allowing it to be applied not only to object recognition but also FER.
Highlights
1 Introduction During the past decade or two, significant effort has been put into developing methods of training algorithms for pattern recognition, which is an attractive research subject in the field of computer vision due to the great potential for it to be used in many applications in a variety of fields, including object recognition, biological feature recognition, and human behavior analysis
Our approach outperforms the methods advocated in a recent line of papers that use third-party software tools to obtain mirror images of samples for training in their object/facial-expression recognition systems, which we briefly review here
Our approach is the first to employ the perturbed subspace method (PSM) directly for detector training without using any tools. Experiments show that this has the greatest impact on the performance of training efficiency, because time can be saved which would otherwise be spent on collecting vast amounts of data from the Internet or using third-party software to deal with the samples in order to get mirror images of three-dimensional data
Summary
During the past decade or two, significant effort has been put into developing methods of training algorithms for pattern recognition, which is an attractive research subject in the field of computer vision due to the great potential for it to be used in many applications in a variety of fields, including object recognition, biological feature recognition, and human behavior analysis. Our approach outperforms the methods advocated in a recent line of papers that use third-party software tools to obtain mirror images of samples for training in their object/facial-expression recognition systems, which we briefly review here. Our approach is the first to employ the PSM directly for detector training without using any tools Experiments show that this has the greatest impact on the performance of training efficiency, because time can be saved which would otherwise be spent on collecting vast amounts of data from the Internet or using third-party software to deal with the samples in order to get mirror images of three-dimensional data.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
