A Multi-Modal Egocentric Activity Recognition Approach towards Video Domain Generalization

Egocentric activity recognition is a prominent computer vision task that is based on the use of wearable cameras. Since egocentric videos are captured through the perspective of the person wearing the camera, her/his body motions severely complicate the video content, imposing several challenges. In this work we propose a novel approach for domain-generalized egocentric human activity recognition. Typical approaches use a large amount of training data, aiming to cover all possible variants of each action. Moreover, several recent approaches have attempted to handle discrepancies between domains with a variety of costly and mostly unsupervised domain adaptation methods. In our approach we show that through simple manipulation of available source domain data and with minor involvement from the target domain, we are able to produce robust models, able to adequately predict human activity in egocentric video sequences. To this end, we introduce a novel three-stream deep neural network architecture combining elements of vision transformers and residual neural networks which are trained using multi-modal data. We evaluate the proposed approach using a challenging, egocentric video dataset and demonstrate its superiority over recent, state-of-the-art research works.

A Deep Regression Approach for Human Activity Recognition Under Partial Occlusion

In real-life scenarios, Human Activity Recognition (HAR) from video data is prone to occlusion of one or more body parts of the human subjects involved. Although it is common sense that the recognition of the majority of activities strongly depends on the motion of some body parts, which when occluded compromise the performance of recognition approaches, this problem is often underestimated in contemporary research works. Currently, training and evaluation is based on datasets that have been shot under laboratory (ideal) conditions, i.e. without any kind of occlusion. In this work, we propose an approach for HAR in the presence of partial occlusion, in cases wherein up to two body parts are involved. We assume that human motion is modeled using a set of 3D skeletal joints and also that occluded body parts remain occluded during the whole duration of the activity. We solve this problem using regression, performed by a novel deep Convolutional Recurrent Neural Network (CRNN). Specifically, given a partially occluded skeleton, we attempt to reconstruct the missing information regarding the motion of its occluded part(s). We evaluate our approach using four publicly available human motion datasets. Our experimental results indicate a significant increase of performance, when compared to baseline approaches, wherein networks that have been trained using only nonoccluded or both occluded and nonoccluded samples are evaluated using occluded samples. To the best of our knowledge, this is the first research work that formulates and copes with the problem of HAR under occlusion as a regression task.

Human Activity Recognition in the Presence of Occlusion

The presence of occlusion in human activity recognition (HAR) tasks hinders the performance of recognition algorithms, as it is responsible for the loss of crucial motion data. Although it is intuitive that it may occur in almost any real-life environment, it is often underestimated in most research works, which tend to rely on datasets that have been collected under ideal conditions, i.e., without any occlusion. In this work, we present an approach that aimed to deal with occlusion in an HAR task. We relied on previous work on HAR and artificially created occluded data samples, assuming that occlusion may prevent the recognition of one or two body parts. The HAR approach we used is based on a Convolutional Neural Network (CNN) that has been trained using 2D representations of 3D skeletal motion. We considered cases in which the network was trained with and without occluded samples and evaluated our approach in single-view, cross-view, and cross-subject cases and using two large scale human motion datasets. Our experimental results indicate that the proposed training strategy is able to provide a significant boost of performance in the presence of occlusion.

A Multimodal Fusion Approach for Human Activity Recognition

The problem of human activity recognition (HAR) has been increasingly attracting the efforts of the research community, having several applications. It consists of recognizing human motion and/or behavior within a given image or a video sequence, using as input raw sensor measurements. In this paper, a multimodal approach addressing the task of video-based HAR is proposed. It is based on 3D visual data that are collected using an RGB + depth camera, resulting to both raw video and 3D skeletal sequences. These data are transformed into six different 2D image representations; four of them are in the spectral domain, another is a pseudo-colored image. The aforementioned representations are based on skeletal data. The last representation is a "dynamic" image which is actually an artificially created image that summarizes RGB data of the whole video sequence, in a visually comprehensible way. In order to classify a given activity video, first, all the aforementioned 2D images are extracted and then six trained convolutional neural networks are used so as to extract visual features. The latter are fused so as to form a single feature vector and are fed into a support vector machine for classification into human activities. For evaluation purposes, a challenging motion activity recognition dataset is used, while single-view, cross-view and cross-subject experiments are performed. Moreover, the proposed approach is compared to three other state-of-the-art methods, demonstrating superior performance in most experiments.

Recognizing Emotional States Using Speech Information

Emotion recognition plays an important role in several applications, such as human computer interaction and understanding affective state of users in certain tasks, e.g., within a learning process, monitoring of elderly, interactive entertainment etc. It may be based upon several modalities, e.g., by analyzing facial expressions and/or speech, using electroencephalograms, electrocardiograms etc. In certain applications the only available modality is the user's (speaker's) voice. In this paper we aim to analyze speakers' emotions based solely on paralinguistic information, i.e., not depending on the linguistic aspect of speech. We compare two machine learning approaches, namely a Convolutional Neural Network and a Support Vector Machine. The former is trained using raw speech information, while the latter is trained on a set of extracted low-level features. Aiming to provide a multilingual approach, training and testing datasets contain speech from different languages.

Meta-analysis: vitamin D and non-alcoholic fatty liver disease

BACKGROUND

Non-alcoholic fatty liver disease (NAFLD) is a highly prevalent condition. Emerging evidence suggests that vitamin D may play a role in the pathogenesis of NAFLD.

AIM

To review systematically the association between vitamin D levels, measured as serum 25-hydroxy vitamin D [25(OH)D], and NAFLD.

METHODS

We used PubMed and EMBASE databases to identify all studies that assessed the association between vitamin D and NAFLD up until 22 April 2013, without language restrictions. We included studies that compared vitamin D levels between NAFLD cases and controls and also those that compared the odds of vitamin D deficiency by NAFLD status. Pooled standardised differences and odds ratios were calculated using an inverse variance method.

RESULTS

Seventeen cross-sectional and case-control studies have evaluated the association between vitamin D and NAFLD. NAFLD was diagnosed using biopsy (4 studies), ultrasound or CT (10 studies) and liver enzymes (3 studies). Nine studies provided data for a quantitative meta-analysis. Compared to controls, NAFLD patients had 0.36 ng/mL (95% CI: 0.32, 0.40 ng/mL) lower levels of 25(OH)D and were 1.26 times more likely to be vitamin D deficient (OR 1.26, 95% CI: 1.17, 1.35).

CONCLUSIONS

NAFLD patients have decreased serum 25(OH)D concentrations, suggesting that vitamin D may play a role in the development of NAFLD. The directionality of this association cannot be determined from cross-sectional studies. Demonstration of a causal role of hypovitaminosis D in NAFLD development in future studies could have important therapeutic implications.

Glycosylation of fluoroquinolones through direct and oxygenated polymethylene linkages as a sugar-mediated active transport system for antimicrobials

We report herein the synthesis and biological testing of several glycosylated derivatives of some fluoroquinolone antibiotics. In particular, we have prepared several glycosylated derivatives of ciprofloxacin (2) in which the carbohydrate units are linked to the free secondary amine of the piperazine unit by: (a) no linker (e.g., a glycosylamine), (b) a beta-oxyethyl linker, and (c) a gamma-oxypropyl linker. Both glucose and galactose were used as carbohydrates so that six compounds of this type were prepared, e.g., no linker 4a,b, oxyethyl linker 5a,b, and oxypropyl linker 6a,b. In addition the aryl glycosides of glucose and galactose (7a,b) were prepared from the active 1-(4-hydroxyphenyl)fluoroquinolone (3.) The syntheses of the glycosylamines 4a,b involved the direct condensation of glucose and galactose with the hydrochloride salt of ciprofloxacin (2). For the oxyalkyl-linked compounds, we first prepared the peracetylated omega-bromoalkyl glycopyranosides 14a,b and 15a,b and then coupled them to the allyl ester of ciprofloxacin (11) to give, after saponification to remove all of the esters, the desired fluoroquinolone carbohydrates 5a,b and 6a,b. The final series was prepared from 2,4,5-trifluorobenzoyl chloride (22) which gave 3 in four precedented steps. Coupling of 3 with the peracetylated glucosyl and galactosyl halides 12a,b and 26 afforded, after saponification, the desired aryl glycosides 7a,b. Six of these derivatives of ciprofloxacin-4a,b, 5a,b, and 6a,b-were subjected to microbiological screening. Of the six, compound 6a showed the highest activity. Since 6a would give the hydroxypropyl-substituted ciprofloxacin on hydrolysis and its activity is approximately 4-8 times less than that of ciprofloxacin (2), this implies that compound 6a is probably being actively transported. Thus preliminary results suggest that some of the compounds are stable in culture conditions and may be differentially transported by multiple resistant organisms. In some cases, the addition of a linker and a carbohydrate to ciprofloxacin lessens, but does not eliminate, antimicrobial activity.

Stimulus-secretion coupling in rat parietal cells is affected by extracellular magnesium

We investigated the effects of extracellular magnesium on stimulus-secretion coupling in isolated rat parietal cells. A high concentration of extracellular magnesium (10 mM) decreased basal and carbachol-stimulated calcium levels, whereas low magnesium levels (0.2 mM) had the opposite effect. The calcium-triggered acid secretion was influenced in the same manner. Basal and carbachol-stimulated acid secretion could be enhanced by incubation in a buffer of low magnesium concentration, whereas a high magnesium concentration totally suppressed the carbachol-induced acid secretion. These results demonstrate that magnesium plays a modulative role in calcium-dependent stimulus-secretion coupling in rat parietal cells.