Edge-assisted Collaborative Image Recognition for Mobile Augmented Reality

Mobile Augmented Reality (AR), which overlays digital content on the real-world scenes surrounding a user, is bringing immersive interactive experiences where the real and virtual worlds are tightly coupled. To enable seamless and precise AR experiences, an image recognition system that can accurately recognize the object in the camera view with low system latency is required. However, due to the pervasiveness and severity of image distortions, an effective and robust image recognition solution for “in the wild” mobile AR is still elusive. In this article, we present CollabAR, an edge-assisted system that provides distortion-tolerant image recognition for mobile AR with imperceptible system latency . CollabAR incorporates both distortion-tolerant and collaborative image recognition modules in its design. The former enables distortion-adaptive image recognition to improve the robustness against image distortions, while the latter exploits the spatial-temporal correlation among mobile AR users to improve recognition accuracy. Moreover, as it is difficult to collect a large-scale image distortion dataset, we propose a Cycle-Consistent Generative Adversarial Network-based data augmentation method to synthesize realistic image distortion. Our evaluation demonstrates that CollabAR achieves over 85% recognition accuracy for “in the wild” images with severe distortions, while reducing the end-to-end system latency to as low as 18.2 ms.

Bug Localization in Model-Based Systems in the Wild

The companies that have adopted the Model-Driven Engineering (MDE) paradigm have the advantage of working at a high level of abstraction. Nevertheless, they have the disadvantage of the lack of tools available to perform bug localization at the model level. In addition, in an MDE context, a bug can be related to different MDE artefacts, such as design-time models, model transformations, or run-time models. Starting the bug localization in the wrong place or with the wrong tool can lead to a result that is unsatisfactory. We evaluate how to apply the existing model-based approaches in order to mitigate the effect of starting the localization in the wrong place. We also take into account that software engineers can refine the results at different stages. In our evaluation, we compare different combinations of the application of bug localization approaches and human refinement. The combination of our approaches plus manual refinement obtains the best results. We performed a statistical analysis to provide evidence of the significance of the results. The conclusions obtained from this evaluation are: humans have to be involved at the right time in the process (or results can even get worse), and artefact-independence can be achieved without worsening the results.

Causes of admission and outcomes of brown hare (Lepus europaeus) leverets at wildlife rescue centres in the Czech Republic

Background Wildlife rescue centres care for orphaned and injured young as an integral part of their work. However, inappropriate interventions in nature can have a negative effect on the survival of young hares, especially when the care of these young is not very successful. The aim of this study was to assess the number of brown hare leverets admitted to rescue centres in the Czech Republic in the period from 2010 to 2019, the causes of their admission to rescue centres and their outcomes. Results We evaluated the number of brown hare leverets admitted to rescue centres in the Czech Republic in the period from 2010 to 2019 and the outcomes associated with their leaving these rescue centres. We found that the number of brown hare leverets admitted increased during the monitored period (rSp = 0.6364, p < 0.05). The most frequent reasons for admission were the admission of orphaned young (49.15%), leverets brought needlessly (19.60%) and leverets that had been bitten by other animals (18.63%). More (p < 0.05) young admitted to rescue centres died (40.76%) than were reared successfully and released back into the wild (32.40%). Leverets that had been caught needlessly or orphaned and late-born leverets survived and could be released back into the wild (38.56, 34.51 and 52%, respectively), while fatalities were recorded in most leverets bitten by another animal (65.05%) or hit in a collision with a vehicle (97.06%). Most young hares (76.92%) that were exhausted or starved at the time of admission could not be saved. Conclusions Since only a small proportion of hares in a litter survive until adulthood in the wild, young animals being found and taken needlessly to rescue centres may harm the hare population. Our results show that only around one in three healthy young hares admitted to rescue centres are reared successfully. It is, in our opinion, of fundamental importance to the protection of brown hare leverets to inform the public of this issue and prevent needless interventions into natural rearing in the wild.

Deciphering the evolution of microbial interactions: in silico studies of two-member microbial communities

Microbes thrive in communities, embedded in a complex web of interactions. These interactions, particularly metabolic interactions, play a crucial role in maintaining the community structure and function. As the organisms thrive and evolve, a variety of evolutionary processes alter the interactions among the organisms in the community, although the community function remains intact. In this work, we simulate the evolution of two-member microbial communities in silico to study how evolutionary forces can shape the interactions between organisms. We employ genomescale metabolic models of organisms from the human gut, which exhibit a range of interaction patterns, from mutualism to parasitism. We observe that the evolution of microbial interactions varies depending upon the starting interaction and also on the metabolic capabilities of the organisms in the community. We find that evolutionary constraints play a significant role in shaping the dependencies of organisms in the community. Evolution of microbial communities yields fitness benefits in only a small fraction of the communities, and is also dependent on the interaction type of the wild-type communities. The metabolites cross-fed in the wild-type communities appear in only less than 50% of the evolved communities. A wide range of new metabolites are cross-fed as the communities evolve. Further, the dynamics of microbial interactions are not specific to the interaction of the wild-type community but vary depending on the organisms present in the community. Our approach of evolving microbial communities in silico provides an exciting glimpse of the dynamics of microbial interactions and offers several avenues for future investigations.

The Habituation Process in Two Groups of Wild Moor Macaques (Macaca maura)

When studying animal behavior in the wild, some behaviors may require observation from a relatively short distance. In these cases, habituation is commonly used to ensure that animals do not perceive researchers as a direct threat and do not alter their behavior in their presence. However, habituation can have significant effects on the welfare and conservation of the animals. Studying how nonhuman primates react to the process of habituation can help to identify the factors that affect habituation and implement habituation protocols that allow other researchers to speed up the process while maintaining high standards of health and safety for both animals and researchers. In this study, we systematically described the habituation of two groups of wild moor macaques (Macaca maura), an Endangered endemic species of Sulawesi Island (Indonesia), to assess the factors that facilitate habituation and reduce impact on animal behavior during this process. During 7 months, we conducted behavioral observations for more than 7,872 encounters and an average of 120 days to monitor how macaque behavior toward researchers changed through time in the two groups under different conditions. We found that both study groups (N = 56, N = 41) became more tolerant to the presence of researchers during the course of the habituation, with occurrence of neutral group responses increasing, and minimum distance to researchers and occurrence of fearful group responses decreasing through time. These changes in behavior were predominant when macaques were in trees, with better visibility conditions, when researchers maintained a longer minimum distance to macaques and, unexpectedly, by the presence of more than one researcher. By identifying these factors, we contribute to designing habituation protocols that decrease the likelihood of fearful responses and might reduce the stress experienced during this process.

Anemone pindariensis (Ranunculaceae), most recently described from the western Himalaya in India, is a teratological form of A. rivularis

Based on literature survey and observations on herbarium specimens and living plants in the wild, we demonstrate that Anemone pindariensis (Ranunculaceae), most recently described from the western Himalaya in India, is a teratological form of A. rivularis. We therefore place the former in synonymy with the latter herein.

The role of the epidermis enhancer element in positive and negative transcriptional regulation of ebony in Drosophila melanogaster

The spatiotemporal regulation of gene expression is essential to ensure robust phenotypic outcomes. Pigmentation patterns in Drosophila are determined by pigments biosynthesized in the developing epidermis and the cis-regulatory elements (CREs) of the genes involved in this process are well-characterized. Here we report that the known primary epidermal enhancer (priEE) is dispensable for the transcriptional activation of ebony (involved in light-colored pigment synthesis) in the developing epidermis of D. melanogaster. The evidence was obtained by introducing an approximately 1 kbp deletion at the priEE by genome editing. The effect of the priEE deletion on pigmentation and on the endogenous expression pattern of a mCherry-fused ebony allele was examined in the abdomen. The expression levels of the mCherry-fused ebony in the priEE-deleted strains were slightly higher than that of the control strain, indicating that the sequences outside the priEE have an ability to drive an expression of this gene in the epidermis. Interestingly, the priEE deletion resulted in a derepression of this gene in the dorsal midline of the abdominal tergites, where dark pigmentation is present in the wild-type individuals. This indicated that the priEE fragment contains a silencer. Furthermore, the endogenous expression pattern of ebony in the two additional strains with partially deleted priEE revealed that the silencer resides within a 351-bp fragment in the 5' portion of the priEE. These results demonstrated that deletion assays combined with reporter assays are highly effective in detecting the presence of positively and negatively regulating sequences within and outside the focal CREs.

Isolation, Identification, and Whole-Genome Sequence Analysis of Pantoea Agglomerans Causing Leaf Spot-Hole Disease in Wild Apricot in Xinjiang, China

Wild apricot in Yili wild fruit forest in Xinjiang have been seriously affected by leaf spot-hole disease, with the incidence reaching 100%. To identify the pathogen of apricot perforation in the Yili wild fruit forest, two bacterial strains with strong virulence were obtained by the dilution separation method. The bacterial strains were gram-negative bacteria with yellow colonies, smooth surfaces and neat edges. The results of the pathogenicity test showed that the bacteria could cause symptoms of leaf spot-hole disease in wild apricot, similar to the symptoms in the field, and could cause HR in tobacco. Based on the 16S rDNA gene sequence and multilocus sequence analysis of fusA, gyrB, leuS, pyrG, rpoB and rlpB, combined with the physiological and biochemical characteristics, the isolated strain was identified as Pantoea agglomerans. The pathogen causing bacterial leaf spot-hole disease in wild apricot was determined to be P. agglomerans in the wild fruit forest of Yili, Xinjiang. The whole genome of the pathogen strain GL9-2 was sequenced based on the Illumina HiSeq500 and PacBio RS platforms. The genome size was 4765392 bp, and the G+C value was 55.27%. There was one chromosome and two plasmids in the genome, and 4353 CDs were identified. The annotation results showed that 52 glycoside hydrolase-related genes, 38 bacterial secretory system-related genes and 600 toxin-related genes were predicted.

Daily Living Activity Recognition In-The-Wild: Modeling and Inferring Activity-Aware Human Contexts

Advancement in smart sensing and computing technologies has provided a dynamic opportunity to develop intelligent systems for human activity monitoring and thus assisted living. Consequently, many researchers have put their efforts into implementing sensor-based activity recognition systems. However, recognizing people’s natural behavior and physical activities with diverse contexts is still a challenging problem because human physical activities are often distracted by changes in their surroundings/environments. Therefore, in addition to physical activity recognition, it is also vital to model and infer the user’s context information to realize human-environment interactions in a better way. Therefore, this research paper proposes a new idea for activity recognition in-the-wild, which entails modeling and identifying detailed human contexts (such as human activities, behavioral environments, and phone states) using portable accelerometer sensors. The proposed scheme offers a detailed/fine-grained representation of natural human activities with contexts, which is crucial for modeling human-environment interactions in context-aware applications/systems effectively. The proposed idea is validated using a series of experiments, and it achieved an average balanced accuracy of 89.43%, which proves its effectiveness.