Understanding Tasks Research Articles

Сovid-19 and International Law: General Issues

INTRODUCTION.InDecember2019,anoutbreakofcoronavirusinfection(SARS-CoV-2)transmittedbyairbornedropletswasrecordedinWuhan,China.Inmid-January,theviruswasdetectedinThailand andJapan;inMarch,thecenterofthepandemicmovedtoEurope;inearlyApril,theUnitedStatescameoutontopintermsofthenumberofinfections.Tocombatthevirus,manystateshaveintroducedemergencymeasures,includinglockdowns,socialdistancerequirements,masstesting,etc.Thepandemichasaffectedallspheresofpubliclife,includinginternationalrelationsandinternationallaw.MATERIALSANDMETHODS.Thearticleanalyzestheresponsetothepandemiconthepartofstates,organizationsandthedoctrineofinternationallaw;examinestheinternationallegalaspectsofthepandemic:applicationoftheInternationalHealthRegulations2005,possibleresponsibilityofChinaandotherstates,impactofthepandemiconhumanrights.Theproblemsoflegalregulationinconnectionwiththepandemicaredefinedandthewaysoftheirsolutionaredetermined.Thesubjectofanalysisisthematerialsofforeignlegalpress,firstofall,postsandarticlesontheInternet.Inadditiontothedataofinternationallaw,scientificcategoriesofphilosophy,economicsandpoliticalscienceareused.RESEARCHRESULTS.MajorUNbodieshavereactedtothepandemicwithgeneralstatements.WHOpositioneditselfasaninternationalcenterforthefightagainstthevirusandmaderecommendationsthat,however, werenotimplementedbystateswhichadoptedmorerestrictivemeasures.ThemaindocumentthatguidedWHOistheInternationalHealthRegulations2005(IHR).SomestatesandthemediaaccusedChinaofa belated reaction and withholding information. As a result, the doctrine has discussed the issue of bringing a claim against China to the International Court of Justice. The legal basis for this claim could be the provisions of the IHR, the WHO Constitution and a number of duediligence obligations. The jurisdictional basis for applying to the ICJ could be Art. 75 of the WHO Constitution, and to an arbitration Art. 56 IHR. In response to the pandemic, many states have limited human rights; references have been made to the possibility of a temporary derogation from human rights obligations in an emergency and the possibility of limiting human rights in the interests of national security, health and the protection of the rights of others.DISCUSSIONANDCONCLUSIONS.Theforeigndoctrinenotesan nsufficient response frominternational organizations and makes proposals aimed at expandingthepowersoftheUNSecurityCouncilincombatingthepandemicandatreformingtheWHO.Inaddition,thereareshortcomingsoftheIHRthathindertheireffectiveuse.ThepossibilityofholdingChinaaccountableisquestioned:thereisnsufficient evidenceoftheviolation; the threshold for breach of duediligence obligationsisveryhigh;Chinaisunlikelytoagreetoparticipateinthelawsuitagainstit. Nevertheless, several lessons canbelearnedforthelawofresponsibility(thepossibilityofdeviatingfromtheprincipleoffullcompensation,etc.).Theprocedureforderogatingfromhumanrightsobligationsduringa pandemic also needs clarification. In general, the international legal doctrine coped with the task of understanding of the pandemic phenomenon: it systematized and qualified the facts, discovered and formulated legal problems, both private and public, and suggested means to solve them. Few questioned the advisability of such a harsh global reaction and formulated a radical criticism; instead, the shortcomings of individual measures were noted and proposals were made to improve their effectiveness.

Business process and rule integration approaches—An empirical analysis of model understanding

Business process models are widely used in organizations by information systems analysts to represent complex business requirements. They are also used by business users to understand business operations and constraints. This understanding is extracted from graphical process models as well as business rules. Prior research advocated integrating business rules and business process models to improve the effectiveness of various organizational activities, such as developing a shared understanding of practices, process improvement, and mitigating risks of compliance and policy breaches. However, whether such integrated modeling can improve the understanding of business processes, which is a fundamental benefit of integrated modeling, has not been empirically evaluated. In this paper, first, we report on an experiment investigating whether rule linking, a representative integrated modeling method, can improve understanding performance. We use eye tracking technology to understand the cognitive process by which model readers use models to perform understanding tasks. Our results show that rule linking outperforms separated modeling in terms of understanding effectiveness, efficiency, perceived mental effort, and visual attention. Further, cognitive process analysis reveals that the form of rule representation does not affect the extent of deep processing, but rule linking significantly decreases the occurrence of rule scanning and screening processes. Moreover, our results show that rule linking leads to an increase of visual association suggesting improved information integration, leading to improved task performance.

Unsupervised Image Segmentation by Mutual Information Maximization and Adversarial Regularization

Semantic segmentation is one of the basic, yet essential scene understanding tasks for an autonomous agent. The recent developments in supervised machine learning and neural networks have enjoyed great success in enhancing the performance of the state-of-the-art techniques for this task. However, their superior performance is highly reliant on the availability of a large-scale annotated dataset. In this letter, we propose a novel fully unsupervised semantic segmentation method, the so-called Information Maximization and Adversarial Regularization Segmentation (InMARS). Inspired by human perception which parses a scene into perceptual groups, rather than analyzing each pixel individually, our proposed approach first partitions an input image into meaningful regions (also known as superpixels). Next, it utilizes Mutual-Information-Maximization followed by an adversarial training strategy to cluster these regions into semantically meaningful classes. To customize an adversarial training scheme for the problem, we incorporate adversarial pixel noise along with spatial perturbations to impose photometrical and geometrical invariance on the deep neural network. Our experiments demonstrate that our method achieves the state-of-the-art performance on two commonly used unsupervised semantic segmentation datasets, COCO-Stuff, and Potsdam.

OryzaGP 2021 update: a rice gene and protein dataset for named-entity recognition.

Due to the rapid evolution of high-throughput technologies, a tremendous amount of data is being produced in the biological domain, which poses a challenging task for information extraction and natural language understanding. Biological named entity recognition (NER) and named entity normalisation (NEN) are two common tasks aiming at identifying and linking biologically important entities such as genes or gene products mentioned in the literature to biological databases. In this paper, we present an updated version of OryzaGP, a gene and protein dataset for rice species created to help natural language processing (NLP) tools in processing NER and NEN tasks. To create the dataset, we selected more than 15,000 abstracts associated with articles previously curated for rice genes. We developed four dictionaries of gene and protein names associated with database identifiers. We used these dictionaries to annotate the dataset. We also annotated the dataset using pre-trained NLP models. Finally, we analysed the annotation results and discussed how to improve OryzaGP.

Collaborative Local-Global Learning for Temporal Action Proposal

Temporal action proposal generation is an essential and challenging task in video understanding, which aims to locate the temporal intervals that likely contain the actions of interest. Although great progress has been made, the problem is still far from being well solved. In particular, prevalent methods can handle well only the local dependencies (i.e., short-term dependencies) among adjacent frames but are generally powerless in dealing with the global dependencies (i.e., long-term dependencies) between distant frames. To tackle this issue, we propose CLGNet, a novel Collaborative Local-Global Learning Network for temporal action proposal. The majority of CLGNet is an integration of Temporal Convolution Network and Bidirectional Long Short-Term Memory, in which Temporal Convolution Network is responsible for local dependencies while Bidirectional Long Short-Term Memory takes charge of handling the global dependencies. Furthermore, an attention mechanism called the background suppression module is designed to guide our model to focus more on the actions. Extensive experiments on two benchmark datasets, THUMOS’14 and ActivityNet-1.3, show that the proposed method can outperform state-of-the-art methods, demonstrating the strong capability of modeling the actions with varying temporal durations.

Post-Institutionalism in the XXI Century: Expanding, Experimenting, Philosophizing

Post-institutionalism is a promising direction in the study of institutions, developing the methodological ideas of critical institutionalism to build an extended institutional approach (in G. Hodgson's terminology). The mission of post-institutionalism is the development of interdisciplinary, complexity-centered methodologies for the analysis of institutions, allowing the development of institutional research beyond the framework of both new and original institutional theories. The article briefly outlines the logic of the creation and origins of the post-institutional theory, provides its methodological features, philosophical foundations, and guidelines for the research program. Post-institutionalists proceed from the fact that the methodological tools of both the new institutional economics and the traditional (“old”) institutionalism are inadequate to the tasks of understanding and explaining the qualitatively complicated institutions of late capitalism. Such institutions are internally heterogeneous, highly fluid, combine different coordinating principles (logics), their functions and boundaries are difficult to identify. The focus of special attention in post-institutional economics is assemblages – ​institutional systems that combine heterogeneous institutions with irreducible logics. Institutional assemblages are highly adaptive but also functionally redundant and conflict-prone. Bricolage is considered as the main type of institutional change in post-institutionalism, which is understood as the recombinant creation of institutions by a multitude of actors from the elements available in the access to solve current institutional problems. Institutional change agents are not only institutional entrepreneurs, but also institutional “workers”, i. e. ordinary actors in their daily routine. The main function of institutions from the point of view of post-institutionalism is not the minimization of transaction costs, but the creation of transaction value.

EVALUATION OF PROGRAMS IN A PAUD CLASSROOM (Study on Lebong Regency SPNF (SKB)

This study aims to reveal, implement, achieve success and the obstacles that affect the success of the tiered training program at the basic level of Early Childhood Educators who have high school qualifications and SI are not the PAUD field in Lebong Regency SPNF (SKB). This research is an evaluation research using Context, Input, Process, Product. The subject of this study was a level-level tiered training participant. Data collection is done by interview method, documentation, and observation. Furthermore, the data obtained were analyzed using analysis at Lebong Regency SPNF (SKB) to obtain data validity, expert discussion, and extension of observation. The results of the study are as follows, context aspect shows the suitability between training participants' needs and participation, training parti1 cipants' experience, and environmental conditions with training program activities, input aspects indicate training participants' motivation, training participant characteristics, resource characteristics, funding, and infrastructure in the category Good, the process aspect shows the training participants' activities, learning strategies and interpersonal relationships are also in good categories. Product aspects show that all levels of basic education and training activities can be implemented well. The success of the basic level education and training program is shown by the improvement in the minimum competence of PAUD educators which includes understanding tasks and authority in assisting educators and accompanying educators. Constraints in the implementation of the training program include less than optimal cooperation, and the media and teaching aids cannot function optimally.

Impact of the Mutational Landscape of the Sodium/Iodide Symporter in Congenital Hypothyroidism.

Background: Iodide transport defect is an uncommon cause of dyshormonogenic congenital hypothyroidism due to homozygous or compound heterozygous pathogenic variants in the SLC5A5 gene, which encodes the sodium/iodide symporter (NIS), causing deficient iodide accumulation in thyroid follicular cells, thus impairing thyroid hormonogenesis. Methods:SLC5A5 gene variants were compiled from public databases and research articles exploring the molecular bases of congenital hypothyroidism. Using a dataset of 198 missense NIS variants classified as either benign or pathogenic, we developed and validated a machine learning-based NIS-specific variant classifier to predict the impact of missense NIS variants. Results: We generated a manually curated dataset containing 7793 unique SLC5A5 variants. As most databases compiled exome sequencing data, variant mapping revealed an increased density of variants in SLC5A5 coding exons. Based on allele frequency (AF) analysis, we established an AF threshold of 1:10,000 above which a variant should be considered benign. Most pathogenic NIS variants were located in the protein-coding region, as most patients were genetically diagnosed by using a candidate gene strategy limited to this region. Significantly, we evidenced that 94.5% of missense NIS variants were classified as of uncertain significance. Therefore, we developed an NIS-specific variant classifier to improve the prediction of pathogenicity of missense variants. Our classifier predicted the clinical outcome of missense variants with high accuracy (90%), outperforming state-of-the-art pathogenicity predictors, such as REVEL, PolyPhen-2, and SIFT. Based on the excellent performance of our classifier, we predicted the mutational landscape of NIS. The analysis of the mutational landscape revealed that most missense variants located in transmembrane segments are frequently pathogenic. Moreover, we predicted that ∼28% of all single-nucleotide variants that could cause missense NIS variants are pathogenic, thus putatively leading to congenital hypothyroidism if present in homozygous or compound heterozygous state. Conclusions: We reported the first NIS-specific variant classifier aiming at improving the interpretation of missense NIS variants in clinical practice. Deciphering the mutational landscape for every protein involved in thyroid hormonogenesis is a relevant task for a deep understanding of the molecular mechanisms causing dyshormonogenic congenital hypothyroidism.

Visual analytics for nonlinear programming in robot motion planning

Nonlinear programming is a complex methodology where a problem is mathematically expressed in terms of optimality while imposing constraints on feasibility. Such problems are formulated by humans and solved by optimization algorithms. We support domain experts in their challenging tasks of understanding and troubleshooting optimization runs of intricate and high-dimensional nonlinear programs through a visual analytics system. The system was designed for our collaborators’ robot motion planning problems, but is domain agnostic in most parts of the visualizations. It allows for an exploration of the iterative solving process of a nonlinear program through several linked views of the computational process. We give insights into this design study, demonstrate our system for selected real-world cases, and discuss the extension of visualization and visual analytics methods for nonlinear programming.Graphic abstract

Protecting and enhancing willingness to communicate with idiodynamic peer-peer strategy sharing

While a wide range of trait-like and situational willingness to communicate (WTC) factors have been investigated, subsequent pedagogical advice often focuses on teacher-led interventions and does not enable learners themselves to overcome immediate WTC related difficulties. This may be because little is known about how students protect and enhance their own WTC levels. This study, therefore, positions students as peer—peer strategy investigators; learners used idiodynamic methodology to interview a selected peer about strategies they used to protect and enhance their WTC during a classroom discussion. Interviewers shared their WTC-related insights in a follow-up class, providing participants with maximum exposure to the elicited WTC-focused strategies. A key finding was that talk-positive students used a combination of expected behaviours and task-outcome-visualization in combination with ‘resetting’ meta-cognitive strategies to overcome learning difficulties. Furthermore, students reported three approaches to peer—peer strategy sharing: desire for self-improvement, rapprochement with peers, and personal reassurance. In addition to understanding task completion- and communication-difficulties and learning potential WTC-focused strategy resolutions, some individuals reported a change in mind-set and continued to observe their peers' learning-focused behaviours after the study. Thus, this idiodynamic peer—peer approach may be used for ongoing WTC-focused learner development and improved intra-group dynamics.

A hierarchical model for learning to understand head gesture videos

Head gesture videos recorded of a person bear rich information about the individual. Automatically understanding these videos can empower many useful human-centered applications in areas such as smart health, education, work safety and security. To understand a video’s content, low-level head gesture signals carried in the video that capture characteristics of both human postures and motions need to be translated into high-level semantic labels. To meet this aim, we propose a hierarchical model for learning to understand head gesture videos. Given a head gesture video of an arbitrary length, the model first segments the full-length video into multiple short clips for clip-based feature extraction. Multiple base feature extraction procedures are then independently tuned via a set of peripheral learning tasks without consuming any labels of the goal task. These independently derived base features are subsequently aggregated through a multi-task learning framework, coupled with a feature dimensionality reduction module, to optimally learn to accomplish the end video understanding task in an weakly supervised manner, utilizing the limited amount of video labels available of the goal task. Experimental results show that the hierarchical model is superior to multiple state-of-the-art peer methods in tackling versatile video understanding tasks.

Real-Time 3D Multi-Object Detection and Localization Based on Deep Learning for Road and Railway Smart Mobility.

For smart mobility, autonomous vehicles, and advanced driver-assistance systems (ADASs), perception of the environment is an important task in scene analysis and understanding. Better perception of the environment allows for enhanced decision making, which, in turn, enables very high-precision actions. To this end, we introduce in this work a new real-time deep learning approach for 3D multi-object detection for smart mobility not only on roads, but also on railways. To obtain the 3D bounding boxes of the objects, we modified a proven real-time 2D detector, YOLOv3, to predict 3D object localization, object dimensions, and object orientation. Our method has been evaluated on KITTI’s road dataset as well as on our own hybrid virtual road/rail dataset acquired from the video game Grand Theft Auto (GTA) V. The evaluation of our method on these two datasets shows good accuracy, but more importantly that it can be used in real-time conditions, in road and rail traffic environments. Through our experimental results, we also show the importance of the accuracy of prediction of the regions of interest (RoIs) used in the estimation of 3D bounding box parameters.

Improving neural machine translation with latent features feedback

Most state-of-the-art neural machine translation (NMT) models progressively encode feature representation in a bottom-up feed-forward fashion. This traditional encoding mechanism has no guidance from external signals. In computer vision tasks, the feedback connection plays a crucial role, particularly for understanding tasks. In this paper, we propose a simple but effective approach to learn latent feature representations explicitly from input sentences via a latent feature encoder (LFE), which are fed back to an NMT encoder via a top-down feedback mechanism. Through the feedback mechanism, the representations in one layer are influenced by representations of both lower and higher layers, resulting in a more effective encoding mechanism. Besides, to enhance the capability of the LFE in better capturing latent features from the source sentences, we pre-train the LFE via a Denoising Auto-Encoder (DAE) strategy. Experimentation on the large-scale WMT 2014 English-to-German and WMT 2017 Chinese-to-English translation tasks demonstrates that our proposed LFE, either pre-trained with the DAE or not, significantly outperforms the strong baseline.

Urban scene understanding based on semantic and socioeconomic features: From high-resolution remote sensing imagery to multi-source geographic datasets

Scene classification is a means to interpret high-resolution remote sensing (HRS) imagery, to obtain the high-level semantic information, which can provide a reliable reference for urban planning and monitoring. The traditional scene classification methods based on HRS imagery take uniform grid cells as the scene units, thereby missing the geographic boundaries and leading to the mosaicking effect. Therefore, in this paper, the urban scene is defined as a geographic unit bordered by the road network. The task of urban scene understanding is to recognize the socioeconomic or natural semantics of the urban scene. However, due to the complexity of the urban environment, the effectiveness of the traditional scene classification methods is limited on account of three problems when applied in urban scenes: 1) The lack of socioeconomic information in HRS images, with which it is difficult to discriminate diverse urban scenes with similar exteriors. 2) The large discrepancy in the sizes and shapes of urban land parcels affects the scene feature extraction and representation. 3) Urban scene understanding frameworks that can embed various scene classification models have rarely been studied. In this paper, to solve these problems, a universal urban scene understanding framework based on multi-source geographic data (USUMG) is proposed. In the USUMG framework, road network and water channel data from OpenStreetMap (OSM) are used for generating the urban scene units. For each irregular unit, a scene decomposition method based on a morphological skeleton is employed to represent the urban scene unit by unified processing patches. To integrate the different data sources, the high-level semantic features extracted from the HRS imagery and the socioeconomic features extracted from point of interest (POI) data are fused to determine the urban scene category. Finally, the USUMG framework with various scene classification methods was tested in urban districts of Wuhan and Macao in China to verify the universality and feasibility of the proposed framework. The experimental performances are provided in this paper as a benchmark for urban scene understanding based on multi-source geographic data.

Sending or not? A multimodal framework for Danmaku comment prediction

Danmaku is an emerging comment design for videos that allows real-time, interactive comments from viewers. Danmaku increases viewers’ interaction with other viewers and streamers, thereby raising viewers’ loyalty and sense of belonging. Sending Danmaku comments demonstrates a higher degree of viewer involvement than traditional static comments below the videos. Therefore, it is necessary and meaningful to learn about viewers’ preferences by observing their behavior, as this may benefit the platform as well as the streamers. However, research on how the multimodal environment affects viewers’ behavior in sending Danmaku comments is quite limited. To fill this gap, we propose a new dataset and a deep neural network integrating multimodal information to predict whether viewers will send Danmaku comments (Deep Multimodal network for Danmaku Forecasting, DMDF) in order to evaluate the impact of the interaction of textual features, audio features and visual features on the behavior of viewers sending Danmaku comments. A series of experimental results based on a real dataset of 249657 samples from Bilibili (a leading Chinese video streaming Website) demonstrate the effectiveness of the proposed DMDF and the helpfulness of all modalities, especially visual and acoustic features, in behavior forecasting. DMDF with the multimodal squeeze-and-excitation (MSE) module we proposed achieves 90.14% on accuracy and 83.60% on F1-score, and it reveals the extent to which a user-generated video can influence viewers to send Danmaku comments, which helps predict viewers’ online viewing behavior. Furthermore, our model contributes to the current work on the video understanding task.

An investigation on the experimental analysis and MATLAB simulation for Dye-Sensitized Solar Cell

Dye-Sensitized Solar Cells (DSSCs) has been extensively explored as a third generation solar cell due to their simple fabrication procedures, better performance especially under low-light intensities, aesthetic features, etc. DSSCs provide an efficient and simple effectuated technology for prospective energy demands. Computing the experimental data in a high performance platform is really an appreciable task for better understanding and accessibility for photovoltaic technology. In this research paper, we fabricate the proposed DSSC, process the experimental data, as an alternative method we compute the experimental data and obtain I-V and IPCE characteristics by programming in MATLAB.

Transformers-sklearn: a toolkit for medical language understanding with transformer-based models

BackgroundTransformer is an attention-based architecture proven the state-of-the-art model in natural language processing (NLP). To reduce the difficulty of beginning to use transformer-based models in medical language understanding and expand the capability of the scikit-learn toolkit in deep learning, we proposed an easy to learn Python toolkit named transformers-sklearn. By wrapping the interfaces of transformers in only three functions (i.e., fit, score, and predict), transformers-sklearn combines the advantages of the transformers and scikit-learn toolkits.MethodsIn transformers-sklearn, three Python classes were implemented, namely, BERTologyClassifier for the classification task, BERTologyNERClassifier for the named entity recognition (NER) task, and BERTologyRegressor for the regression task. Each class contains three methods, i.e., fit for fine-tuning transformer-based models with the training dataset, score for evaluating the performance of the fine-tuned model, and predict for predicting the labels of the test dataset. transformers-sklearn is a user-friendly toolkit that (1) Is customizable via a few parameters (e.g., model_name_or_path and model_type), (2) Supports multilingual NLP tasks, and (3) Requires less coding. The input data format is automatically generated by transformers-sklearn with the annotated corpus. Newcomers only need to prepare the dataset. The model framework and training methods are predefined in transformers-sklearn.ResultsWe collected four open-source medical language datasets, including TrialClassification for Chinese medical trial text multi label classification, BC5CDR for English biomedical text name entity recognition, DiabetesNER for Chinese diabetes entity recognition and BIOSSES for English biomedical sentence similarity estimation.In the four medical NLP tasks, the average code size of our script is 45 lines/task, which is one-sixth the size of transformers’ script. The experimental results show that transformers-sklearn based on pretrained BERT models achieved macro F1 scores of 0.8225, 0.8703 and 0.6908, respectively, on the TrialClassification, BC5CDR and DiabetesNER tasks and a Pearson correlation of 0.8260 on the BIOSSES task, which is consistent with the results of transformers.ConclusionsThe proposed toolkit could help newcomers address medical language understanding tasks using the scikit-learn coding style easily. The code and tutorials of transformers-sklearn are available at https://doi.org/10.5281/zenodo.4453803. In future, more medical language understanding tasks will be supported to improve the applications of transformers_sklearn.

Социокультурные и общественно-политические контексты участия мусульманской молодежи в институциональной политике Шотландии

The authors emphasize that with the growth of Scottish national consciousness and the reality of a new independence referendum, Scotland is increasingly developing a special socio-cultural and political landscape that offers young Muslims different contexts of political participation. According to the authors, young Muslims represent a very important demographic group in Scotland. In this regard, the study of the place and role of representatives of Muslim youth in the socio-political processes of Scotland is particularly relevant in scientific, theoretical and practical aspects. The authors of this article, set themselves the task of scientific understanding of the socio-cultural and socio-political contexts of the participation of Muslim youth in the institutional policy of Scotland. They highlight the significant role the SNP plays in engaging Muslim youth in politics. This is possible because the SNP is able to use various socio-political issues of young Muslims to involve the latter in the implementation of the Scottish nationalist agenda actively. The authors conclude that, despite the variety of forms and ways in which young Muslims express their interest in politics, their participation is primarily due to the Scottish (rather than British) specifics of the socio-cultural and socio-political contexts, which are present in this part of the UK.

Image Caption Generation Using Multi-Level Semantic Context Information

Object detection, visual relationship detection, and image captioning, which are the three main visual tasks in scene understanding, are highly correlated and correspond to different semantic levels of scene image. However, the existing captioning methods convert the extracted image features into description text, and the obtained results are not satisfactory. In this work, we propose a Multi-level Semantic Context Information (MSCI) network with an overall symmetrical structure to leverage the mutual connections across the three different semantic layers and extract the context information between them, to solve jointly the three vision tasks for achieving the accurate and comprehensive description of the scene image. The model uses a feature refining structure to mutual connections and iteratively updates the different semantic features of the image. Then a context information extraction network is used to extract the context information between the three different semantic layers, and an attention mechanism is introduced to improve the accuracy of image captioning while using the context information between the different semantic layers to improve the accuracy of object detection and relationship detection. Experiments on the VRD and COCO datasets demonstrate that our proposed model can leverage the context information between semantic layers to improve the accuracy of those visual tasks generation.

Investigating the utility of VR for spatial understanding in surgical planning: evaluation of head-mounted to desktop display

Recent technological advances have made Virtual Reality (VR) attractive in both research and real world applications such as training, rehabilitation, and gaming. Although these other fields benefited from VR technology, it remains unclear whether VR contributes to better spatial understanding and training in the context of surgical planning. In this study, we evaluated the use of VR by comparing the recall of spatial information in two learning conditions: a head-mounted display (HMD) and a desktop screen (DT). Specifically, we explored (a) a scene understanding and then (b) a direction estimation task using two 3D models (i.e., a liver and a pyramid). In the scene understanding task, participants had to navigate the rendered the 3D models by means of rotation, zoom and transparency in order to substantially identify the spatial relationships among its internal objects. In the subsequent direction estimation task, participants had to point at a previously identified target object, i.e., internal sphere, on a materialized 3D-printed version of the model using a tracked pointing tool. Results showed that the learning condition (HMD or DT) did not influence participants’ memory and confidence ratings of the models. In contrast, the model type, that is, whether the model to be recalled was a liver or a pyramid significantly affected participants’ memory about the internal structure of the model. Furthermore, localizing the internal position of the target sphere was also unaffected by participants’ previous experience of the model via HMD or DT. Overall, results provide novel insights on the use of VR in a surgical planning scenario and have paramount implications in medical learning by shedding light on the mental model we make to recall spatial structures.