In crowd behavior studies, a model of crowd behavior needs to be trained using the information extracted from video sequences. Most of the previous methods are based on low-level visual features because there are only crowd behavior labels available as ground-truth information in crowd datasets. However, there is a huge semantic gap between low-level motion/appearance features and high-level concept of crowd behaviors. In this paper, we tackle the problem by introducing an attribute-based scheme. While similar strategies have been employed for action and object recognition, to the best of our knowledge, for the first time it is shown that the crowd emotions can be used as attributes for crowd behavior understanding. We explore the idea of training a set of emotion-based classifiers, which can subsequently be used to indicate the crowd motion. In this scheme, we collect a large dataset of video clips and provide them with both annotations of “crowd behaviors” and “crowd emotions”. We test the proposed emotion based crowd representation methods on our dataset. The obtained promising results demonstrate that the crowd emotions enable the construction of more descriptive models for crowd behaviors. We aim at publishing the dataset with the article, to be used as a benchmark for the communities.
There scientific and therapeutic advances in perinatology and neonatology have improved the survival prospects of preterm and extremely low birth weight infants. Infants' cries are a valuable noninvasive tool for monitoring their neurologic health, especially if they are premature. An automatic acoustic analysis and data mining are employed in this study to determine the discriminative features of preterm and full-term infant cries. The use of machine learning for recognizing sounds in a newborn's cry language has received less attention than previous methods for analyzing the sounds. Moreover, to extract appropriate features from infant cries, adequate knowledge and appropriate signal descriptors are required. Accordingly, to analyze infant cry language, we propose an approach that uses fractal descriptor to extract discriminant features from spectrograms of windowed signals, followed by iterative neighborhood component analysis (iNCA) to select appropriate features. Additionally, the improved deep support vector machine (deepSVM) is used to classify the infants' crying types and their meanings. The proposed method is verified using a newborn sound dataset. According to the classification of five types of crying perception based on various characteristics, 98.34% of all crying perceptions have been recognized. Although there are many classes examined, the feature extraction method based on the fractal method and our optimal classification have a much higher diagnostic accuracy compared to similar methods for analyzing baby crying language. The proposed method can overcome many problems associated with analyzing babies' crying sounds and understanding their language, like uncertainty and unusual errors in classification.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.