Background: Mental health conditions pose a major challenge to health care providers and society at large. The World Health Organization predicts that by 2030, mental health conditions will be the leading cause of disease burden worldwide. The current need for mental health care is overwhelming. In New Zealand, 1 in 6 adults has been diagnosed with common mental disorders, such as depression and anxiety disorders, according to a national survey. Cognitive behavioral therapy (CBT) has been shown to effectively help patients overcome a wide variety of mental health conditions. Virtual reality exposure therapy (VRET) might be one of the most exciting technologies emerging in the clinical setting for the treatment of anxiety and depression. Objective: This study aims to investigate the virtual reality (VR) technologies currently being used to help support the treatment of depression and anxiety. We also aim to investigate whether and how CBT is included as part of VRET and look at the VR technologies and interventions that have been used in recent studies on depression and anxiety. Methods: We performed a scoping review. To identify significant studies, we decided to use already aggregated sources from the Google Scholar database. Overall, the goal of our search strategy was to limit the number of initial results related to VR in mental health to only a relevant minimum. Results: Using our defined keywords, Google Scholar identified >17,300 articles. After applying all the inclusion and exclusion criteria, we identified a total of 369 articles for further processing. After manual evaluation, 34 articles were shortlisted; of the 34 articles, 9 (26%) reported the use of CBT with VR. All of the articles were published between 2017 and 2021. Out of the 9 studies, CBT was conducted within a VR environment in 5 (56%) studies, whereas in the remaining 4 (44%) studies, CBT was used as an addition to VRET. All 9 studies reported the use of CBT either in vivo or in a virtual environment to be effective in supporting the treatment of anxiety or depression. Conclusions: Most studies demonstrated the use of VR to be effective for supporting the treatment of anxiety or depression in a range of settings and recommended its potential as a tool for use in a clinical environment. Even though standalone headsets are much easier to work with and more suitable for home use, the shift from tethered VR headsets to standalone headsets in the mental health environment was not observed. All studies that looked at the use of CBT either in vivo or in a virtual environment found it to be effective in supporting the treatment of anxiety or depression.

The refresh rate of virtual reality (VR) head-mounted displays (HMDs) has been growing rapidly in recent years because of the demand to provide higher frame rate content as it is often linked with a better experience. Today's HMDs come with different refresh rates ranging from 20Hz to 180Hz, which determines the actual maximum frame rate perceived by users' naked eyes. VR users and content developers often face a choice because having high frame rate content and the hardware that supports it comes with higher costs and other trade-offs (such as heavier and bulkier HMDs). Both VR users and developers can choose a suitable frame rate if they are aware of the benefits of different frame rates in user experience, performance, and simulator sickness (SS). To our knowledge, limited research on frame rate in VR HMDs is available. In this paper, we aim to fill this gap and report a study with two VR application scenarios that compared four of the most common and highest frame rates currently available (60, 90, 120, and 180 frames per second (fps)) to explore their effect on users' experience, performance, and SS symptoms. Our results show that 120fps is an important threshold for VR. After 120fps, users tend to feel lower SS symptoms without a significant negative effect on their experience. Higher frame rates (e.g., 120 and 180fps) can ensure better user performance than lower rates. Interestingly, we also found that at 60fps and when users are faced with fast-moving objects, they tend to adopt a strategy to compensate for the lack of visual details by predicting or filling the gaps to try to meet the performance needs. At higher fps, users do not need to follow this compensatory strategy to meet the fast response performance requirements.

Adversarial examples are augmented data points generated by imperceptible perturbation of input samples. They have recently drawn much attention with the machine learning and data mining community. Being difficult to distinguish from real examples, such adversarial examples could change the prediction of many of the best learning models including the state-of-the-art deep learning models. Recent attempts have been made to build robust models that take into account adversarial examples. However, these methods can either lead to performance drops or lack mathematical motivations. In this paper, we propose a unified framework to build robust machine learning models against adversarial examples. More specifically, using the unified framework, we develop a family of gradient regularization methods that effectively penalize the gradient of loss function w.r.t. inputs. Our proposed framework is appealing in that it offers a unified view to deal with adversarial examples. It incorporates another recently-proposed perturbation based approach as a special case. In addition, we present some visual effects that reveals semantic meaning in those perturbations, and thus support our regularization method and provide another explanation for generalizability of adversarial examples. By applying this technique to Maxout networks, we conduct a series of experiments and achieve encouraging results on two benchmark datasets. In particular, we attain the best accuracy on MNIST data (without data augmentation) and competitive performance on CIFAR-10 data.

In real scenes, due to the problems of low light and unsuitable views, the images often exhibit a variety of degradations, such as low contrast, color distortion, and noise. These degradations affect not only visual effects but also computer vision tasks. This paper focuses on the combination of traditional algorithms and machine learning algorithms in the field of image enhancement. The traditional methods, including their principles and improvements, are introduced from three categories: gray level transformation, histogram equalization, and Retinex methods. Machine learning based algorithms are not only divided into end-to-end learning and unpaired learning, but also concluded to decomposition-based learning and fusion based learning based on the applied image processing strategies. Finally, the involved methods are comprehensively compared by multiple image quality assessment methods, including mean square error, natural image quality evaluator, structural similarity, peak signal to noise ratio, etc.

Virtual reality (VR) has advanced rapidly and is used for many entertainment and business purposes. The need for secure, transparent and non-intrusive identification mechanisms is important to facilitate users’ safe participation and secure experience. People are kinesiologically unique, having individual behavioral and movement characteristics, which can be leveraged and used in security sensitive VR applications to compensate for users’ inability to detect potential observational attackers in the physical world. Additionally, such method of identification using a user’s kinesiological data is valuable in common scenarios where multiple users simultaneously participate in a VR environment. In this paper, we present a user study (n = 15) where our participants performed a series of controlled tasks that require physical movements (such as grabbing, rotating and dropping) that could be decomposed into unique kinesiological patterns while we monitored and captured their hand, head and eye gaze data within the VR environment. We present an analysis of the data and show that these data can be used as a biometric discriminant of high confidence using machine learning classification methods such as kNN or SVM, thereby adding a layer of security in terms of identification or dynamically adapting the VR environment to the users’ preferences. We also performed a whitebox penetration testing with 12 attackers, some of whom were physically similar to the participants. We could obtain an average identification confidence value of 0.98 from the actual participants’ test data after the initial study and also a trained model classification accuracy of 98.6%. Penetration testing indicated all attackers resulted in confidence values of less than 50% (<50%), although physically similar attackers had higher confidence values. These findings can help the design and development of secure VR systems.

Understanding the endpoint distribution of pointing selection tasks can reveal the underlying patterns on how users tend to acquire a target, which is one of the most essential and pervasive tasks in interactive systems. It could further aid designers to create new graphical user interfaces and interaction techniques that are optimized for accuracy, efficiency, and ease of use. Previous research has explored the modeling of endpoint distribution outside of virtual reality (VR) systems that have shown to be useful in predicting selection accuracy and guide the design of new interactive techniques. This work aims at developing an endpoint distribution of selection tasks for VR systems which has resulted in EDModel , a novel model that can be used to predict endpoint distribution of pointing selection tasks in VR environments. The development of EDModel is based on two users studies that have explored how factors such as target size, movement amplitude, and target depth affect the endpoint distribution. The model is built from the collected data and its generalizability is subsequently tested in complex scenarios with more relaxed conditions. Three applications of EDModel inspired by previous research are evaluated to show the broad applicability and usefulness of the model: correcting the bias in Fitts's law, predicting selection accuracy, and enhancing pointing selection techniques. Overall, EDModel can achieve high prediction accuracy and can be adapted to different types of applications in VR.