Sequential Variational Autoencoder with Adversarial Classifier for Video Disentanglement.

In this paper, we propose a sequential variational autoencoder for video disentanglement, which is a representation learning method that can be used to separately extract static and dynamic features from videos. Building sequential variational autoencoders with a two-stream architecture induces inductive bias for video disentanglement. However, our preliminary experiment demonstrated that the two-stream architecture is insufficient for video disentanglement because static features frequently contain dynamic features. Additionally, we found that dynamic features are not discriminative in the latent space. To address these problems, we introduced an adversarial classifier using supervised learning into the two-stream architecture. The strong inductive bias through supervision separates dynamic features from static features and yields discriminative representations of the dynamic features. Through a comparison with other sequential variational autoencoders, we qualitatively and quantitatively demonstrate the effectiveness of the proposed method on the Sprites and MUG datasets.

Spatiotemporal forecasting of vertical track alignment with exogenous factors.

To ensure the safety of railroad operations, it is important to monitor and forecast track geometry irregularities. A higher safety requires forecasting with higher spatiotemporal frequencies, which in turn requires capturing spatial correlations. Additionally, track geometry irregularities are influenced by multiple exogenous factors. In this study, a method is proposed to forecast one type of track geometry irregularity, vertical alignment, by incorporating spatial and exogenous factor calculations. The proposed method embeds exogenous factors and captures spatiotemporal correlations using a convolutional long short-term memory. The proposed method is also experimentally compared with other methods in terms of the forecasting performance. Additionally, an ablation study on exogenous factors is conducted to examine their individual contributions to the forecasting performance. The results reveal that spatial calculations and maintenance record data improve the forecasting of vertical alignment.

Framework for identifying and visualising emotional atmosphere in online learning environments in the COVID-19 Era.

In addition to the almost five million lives lost and millions more than that in hospitalisations, efforts to mitigate the spread of the COVID-19 pandemic, which that has disrupted every aspect of human life deserves the contributions of all and sundry. Education is one of the areas most affected by the COVID-imposed abhorrence to physical (i.e., face-to-face (F2F)) communication. Consequently, schools, colleges, and universities worldwide have been forced to transition to different forms of online and virtual learning. Unlike F2F classes where the instructors could monitor and adjust lessons and content in tandem with the learners' perceived emotions and engagement, in online learning environments (OLE), such tasks are daunting to undertake. In our modest contribution to ameliorate disruptions to education caused by the pandemic, this study presents an intuitive model to monitor the concentration, understanding, and engagement expected of a productive classroom environment. The proposed apposite OLE (i.e., AOLE) provides an intelligent 3D visualisation of the classroom atmosphere (CA), which could assist instructors adjust and tailor both content and instruction for maximum delivery. Furthermore, individual learner status could be tracked via visualisation of his/her emotion curve at any stage of the lesson or learning cycle. Considering the enormous emotional and psychological toll caused by COVID and the attendant shift to OLE, the emotion curves could be progressively compared through the duration of the learning cycle and the semester to track learners' performance through to the final examinations. In terms of learning within the CA, our proposed AOLE is assessed within a class of 15 students and three instructors. Correlation of the outcomes reported with those from administered questionnaires validate the potential of our proposed model as a support for learning and counselling during these unprecedentedtimes that we find ourselves.

Zero-Shot Action Recognition with Three-Stream Graph Convolutional Networks.

Large datasets are often used to improve the accuracy of action recognition. However, very large datasets are problematic as, for example, the annotation of large datasets is labor-intensive. This has encouraged research in zero-shot action recognition (ZSAR). Presently, most ZSAR methods recognize actions according to each video frame. These methods are affected by light, camera angle, and background, and most methods are unable to process time series data. The accuracy of the model is reduced owing to these reasons. In this paper, in order to solve these problems, we propose a three-stream graph convolutional network that processes both types of data. Our model has two parts. One part can process RGB data, which contains extensive useful information. The other part can process skeleton data, which is not affected by light and background. By combining these two outputs with a weighted sum, our model predicts the final results for ZSAR. Experiments conducted on three datasets demonstrate that our model has greater accuracy than a baseline model. Moreover, we also prove that our model can learn from human experience, which can make the model more accurate.

2D Convolutional Neural Markov Models for Spatiotemporal Sequence Forecasting.

Recent approaches to time series forecasting, especially forecasting spatiotemporal sequences, have leveraged the approximation power of deep neural networks to model the complexity of such sequences, specifically approaches that are based on recurrent neural networks. Still, as spatiotemporal sequences that arise in the real world are noisy and chaotic, modeling approaches that utilize probabilistic temporal models, such as deep Markov models (DMMs), are favorable because of their ability to model uncertainty, increasing their robustness to noise. However, approaches based on DMMs do not maintain the spatial characteristics of spatiotemporal sequences, with most of the approaches converting the observed input into 1D data halfway through the model. To solve this, we propose a model that retains the spatial aspect of the target sequence with a DMM that consists of 2D convolutional neural networks. We then show the robustness of our method to data with large variance compared with naive forecast, vanilla DMM, and convolutional long short-term memory (LSTM) using synthetic data, even outperforming the DNN models over a longer forecast period. We also point out the limitations of our model when forecasting real-world precipitation data and the possible future work that can be done to address these limitations, along with additional future research potential.

[Anesthetic management for a patient with platypnea-orthodeoxia syndrome].

We experienced anesthetic management for a patient with platypnea-orthodeoxia syndrome. This syndrome is relatively uncommon and accompanies dyspnea and hypoxemia on changing to a sitting or standing from recumbent position. A 75-year-old man with the syndrome underwent atrial septal defect closure on cardiopulmonary bypass. General anesthesia was induced and maintained with midazolam, propofol, fentanyl and vecuronium bromide. During the induction, Spo2 decreased suddenly from 100% to 70%, Spo2, however, recovered to 97% immediately after changing to Trendelenburg position. The perioperative and postoperative course was uneventful, except for hypoxemia during induction. Although the exact mechanisms of platypneaorthodeoxia remains to be solved, right-to-left shunt by an anatomical abnormality and by change of the atrial septum is considered one of the hypoxic mechanisms. We suggest that it is necessary to prevent right-to-left shunt and hypoxemia in anesthetic management of a patient with platypneaorthodeoxia syndrome.