Audio-Visual Emotion Recognition System using Multi-Modal Features

Due to the highly variant face geometry and appearances, Facial Expression Recognition (FER) is still a challenging problem. CNN can characterize 2-D signals. Therefore, for emotion recognition in a video, the authors propose a feature selection model in AlexNet architecture to extract and filter facial features automatically. Similarly, for emotion recognition in audio, the authors use a deep LSTM-RNN. Finally, they propose a probabilistic model for the fusion of audio and visual models using facial features and speech of a subject. The model combines all the extracted features and use them to train the linear SVM (Support Vector Machine) classifiers. The proposed model outperforms the other existing models and achieves state-of-the-art performance for audio, visual and fusion models. The model classifies the seven known facial expressions, namely anger, happy, surprise, fear, disgust, sad, and neutral on the eNTERFACE’05 dataset with an overall accuracy of 76.61%.

Audio-Visual Emotion Recognition System Using Multi-Modal Features

Due to the highly variant face geometry and appearances, Facial Expression Recognition (FER) is still a challenging problem. CNN can characterize 2-D signals. Therefore, for emotion recognition in a video, the authors propose a feature selection model in AlexNet architecture to extract and filter facial features automatically. Similarly, for emotion recognition in audio, the authors use a deep LSTM-RNN. Finally, they propose a probabilistic model for the fusion of audio and visual models using facial features and speech of a subject. The model combines all the extracted features and use them to train the linear SVM (Support Vector Machine) classifiers. The proposed model outperforms the other existing models and achieves state-of-the-art performance for audio, visual and fusion models. The model classifies the seven known facial expressions, namely anger, happy, surprise, fear, disgust, sad, and neutral on the eNTERFACE’05 dataset with an overall accuracy of 76.61%.

Topic-Oriented Text Features Can Match Visual Deep Models of Video Memorability

Not every visual media production is equally retained in memory. Recent studies have shown that the elements of an image, as well as their mutual semantic dependencies, provide a strong clue as to whether a video clip will be recalled on a second viewing or not. We believe that short textual descriptions encapsulate most of these relationships among the elements of a video, and thus they represent a rich yet concise source of information to tackle the problem of media memorability prediction. In this paper, we deepen the study of short captions as a means to convey in natural language the visual semantics of a video. We propose to use vector embeddings from a pretrained SBERT topic detection model with no adaptation as input features to a linear regression model, showing that, from such a representation, simpler algorithms can outperform deep visual models. Our results suggest that text descriptions expressed in natural language might be effective in embodying the visual semantics required to model video memorability.

Web Interoperability for Ontology Development and Support with crowd 2.0

In this work, we treat web interoperability in terms of interchanging ontologies (as knowledge models) within user-centred ontology engineering environments, involving visual and serialised representations of ontologies. To do this, we deal with the tool interoperability problem by re-using an enough expressive ontology-driven metamodel, named KF, proposed as a bridge for interchanging both knowledge models. We provide an extensible web framework, named crowd 2.0, unifying the standard conceptual data modelling languages for generating OWL 2 ontologies from semantic visualisations. Visual models are designed as UML, ER or ORM 2 diagrams, represented as KF instances, and finally, formalised as DL-based models. Reasoning results may be newly incorporated into the shared KF instance to be visualised in any of the provided languages.

The visual ecology of selective predation: Are unhealthy hosts less stealthy hosts?

Predators can strongly influence disease transmission and evolution, particularly when they prey selectively on infected hosts. Although selective predation has been observed in numerous systems, why predators select infected prey remains poorly understood. Here, we use a model of predator vision to test a longstanding hypothesis as to the mechanistic basis of selective predation in a Daphnia-microparasite system, which serves as a model for the ecology and evolution of infectious diseases. Bluegill sunfish feed selectively on Daphnia with a variety of parasites, particularly in water uncolored by dissolved organic carbon. The leading hypothesis for selective predation in this system is that infection-induced changes in the appearance of Daphnia render them more visible to bluegill. Rigorously evaluating this hypothesis requires that we quantify the effect of infection on the visibility of prey from the predator’s perspective, rather than our own. Using a model of the bluegill visual system, we show that the three common parasites, Metschnikowia bicuspidata, Pasteuria ramosa and Spirobacillus cienkowskii, increase the opacity of Daphnia, rendering infected Daphnia darker against the background of downwelling light. As a result of this increased brightness contrast, bluegill can see infected Daphnia at greater distances than uninfected Daphnia – between 19-33% further, depending on the parasite. Pasteuria and Spirobacillus also increase the chromatic contrast of Daphnia. Contrary to expectations, the visibility Daphnia was not strongly impacted by water color in our model. Our work generates hypotheses about which parasites are most likely affected by selective predation in this important model system and establishes visual models as a valuable tool for understanding ecological interactions that impact disease transmission.

Parading the Temporary: Cosmoramas, Panoramas, and Spectacles in Early Nineteenth-Century Paris

The revolution of 1789 prompted various socio-cultural changes that deeply affected French society. Alongside the sense of instability that these events provoked, there are a number of open-air amusements, shows, exhibitions, and theatrical representations, from the Directoire and through the Napoleonic era. This chapter aims to analyze the mechanisms that allowed the development of these spaces. Ephemerality and temporality are central to this investigation, often determining the development of the space, its construction and functions, but also the cultural practices this comprehension of the space engendered. My objective is to discuss the visual models and cultural references enabling the rearrangement of existing areas and the rise of new “spheres” devoted to the consumption of entertainment.

“Conflict of Goals” as a Barrier for Effective Use of Visual Models in Primary Math Education

The formation of the ability to solve non-trivial life problems is one of the tasks of school education in the context of achieving sustainable development goals. In the process of teaching mathematics, one of the most effective ways to find solutions to problems is modelling – a teaching method that not only helps students to consciously assimilate mathematical content, but also forms the basis for selfstudy throughout life. Visual models, which reflect the essential characteristics of mathematical concepts by pictorial means, play a special role in the process of initial teaching of mathematics. Teachers can use active and passive techniques for working with visual models in mathematics lessons, which differ in the degree of children’s participation in building a visual model. The main goal of this article is to identify which techniques teachers prefer working with visual models in practice in mathematics lessons. To achieve this goal, the questionnaire method, the multi-criteria assessment method, and the moderation method were applied. This article presents the results of a study devoted to identifying teachers’ preferred methods of working with visual models when conducting mathematics lessons, identifying their theoretical ideas about the value of each group of techniques, as well as establishing the reasons for the revealed discrepancy between the practical preferences of teachers and their theoretical ideas.