Training and Testing Texture Similarity Metrics for Structurally Lossless Compression.

We present a systematic approach for training and testing structural texture similarity metrics (STSIMs) so that they can be used to exploit texture redundancy for structurally lossless image compression. The training and testing is based on a set of image distortions that reflect the characteristics of the perturbations present in natural texture images. We conduct empirical studies to determine the perceived similarity scale across all pairs of original and distorted textures. We then introduce a data-driven approach for training the Mahalanobis formulation of STSIM based on the resulting annotated texture pairs. Experimental results demonstrate that training results in significant improvements in metric performance. We also show that the performance of the trained STSIM metrics is competitive with state of the art metrics based on convolutional neural networks, at substantially lower computational cost.

Inclusion Depth for Contour Ensembles.

Ensembles of contours arise in various applications like simulation, computer-aided design, and semantic segmentation. Uncovering ensemble patterns and analyzing individual members is a challenging task that suffers from clutter. Ensemble statistical summarization can alleviate this issue by permitting analyzing ensembles' distributional components like the mean and median, confidence intervals, and outliers. Contour boxplots, powered by Contour Band Depth (CBD), are a popular non-parametric ensemble summarization method that benefits from CBD's generality, robustness, and theoretical properties. In this work, we introduce Inclusion Depth (ID), a new notion of contour depth with three defining characteristics. First, ID is a generalization of functional Half-Region Depth, which offers several theoretical guarantees. Second, ID relies on a simple principle: the inside/outside relationships between contours. This facilitates implementing ID and understanding its results. Third, the computational complexity of ID scales quadratically in the number of members of the ensemble, improving CBD's cubic complexity. This also in practice speeds up the computation enabling the use of ID for exploring large contour ensembles or in contexts requiring multiple depth evaluations like clustering. In a series of experiments on synthetic data and case studies with meteorological and segmentation data, we evaluate ID's performance and demonstrate its capabilities for the visual analysis of contour ensembles.

Developments in work-related upper limb disorders (WRULD) amongst Dutch university students from 2004 to 2014.

BACKGROUND: Former studies on work-related upper limb disorders (WRULD) within university education report substantial prevalence rates. In this study, developments in WRULD amongst students in the period 2004-2014 were investigated. Our findings can be a benchmark for future studies, in particular when there are major societal changes as in the case of the COVID-19 pandemic. OBJECTIVE: Differences in time (academic year), how long students have been studying (year of studying), relations with computer time and societal changes were points of interest. METHODS: 2254 students (average age 20.0 years) responded to a questionnaire on WRULD. Students experiencing complaints were further questioned about the severity of complaints and associated body locations. RESULTS: The average percentage of students experiencing complaints was 57%. The highest prevalence rates and severity scores were found in the first and last recorded academic years. The neck, shoulder, back and wrist were most often indicated. The prevalence of complaints raised from the 1st (49%) to the 4th (75%) year of studying. Two seriousness measures showed highest scores in the 5th/6th/7th year of studying. Relations were found between both the prevalence and seriousness of complaints with reported computer time. CONCLUSIONS: After an initial decreasing trend from the academic year 2006/2007 to 2010/2011 there was an increase in WRULD amongst students from 2010/2011 to 2013/2014. Limiting financial and study time factors may have played a role. Structural attention for WRULD prevention and risk factors seems to be effective in reducing prevalence and severity of WRULD. This seems to be even more necessary due to recent COVID-related changes in the students' lives.

Hysteresis in Mental Workload and Task Performance: The Influence of Demand Transitions and Task Prioritization.

OBJECTIVE: We examine how transitions in task demand are manifested in mental workload and performance in a dual-task setting. BACKGROUND: Hysteresis has been defined as the ongoing influence of demand levels prior to a demand transition. Authors of previous studies predominantly examined hysteretic effects in terms of performance. However, little is known about the temporal development of hysteresis in mental workload. METHOD: A simulated driving task was combined with an auditory memory task. Participants were instructed to prioritize driving or to prioritize both tasks equally. Three experimental conditions with low, high, and low task demands were constructed by manipulating the frequency of lane changing. Multiple measures of subjective mental workload were taken during experimental conditions. RESULTS: Contrary to our prediction, no hysteretic effects were found after the high- to low-demand transition. However, a hysteretic effect in mental workload was found within the high-demand condition, which degraded toward the end of the high condition. Priority instructions were not reflected in performance. CONCLUSION: Online assessment of both performance and mental workload demonstrates the transient nature of hysteretic effects. An explanation for the observed hysteretic effect in mental workload is offered in terms of effort regulation. APPLICATION: An informed arrival at the scene is important in safety operations, but peaks in mental workload should be avoided to prevent buildup of fatigue. Therefore, communication technologies should incorporate the historical profile of task demand.

Task Prioritization in Dual-Tasking: Instructions versus Preferences.

The role of task prioritization in performance tradeoffs during multi-tasking has received widespread attention. However, little is known on whether people have preferences regarding tasks, and if so, whether these preferences conflict with priority instructions. Three experiments were conducted with a high-speed driving game and an auditory memory task. In Experiment 1, participants did not receive priority instructions. Participants performed different sequences of single-task and dual-task conditions. Task performance was evaluated according to participants' retrospective accounts on preferences. These preferences were reformulated as priority instructions in Experiments 2 and 3. The results showed that people differ in their preferences regarding task prioritization in an experimental setting, which can be overruled by priority instructions, but only after increased dual-task exposure. Additional measures of mental effort showed that performance tradeoffs had an impact on mental effort. The interpretation of these findings was used to explore an extension of Threaded Cognition Theory with Hockey's Compensatory Control Model.

Effective and efficient subjective testing of texture similarity metrics.

The development and testing of objective texture similarity metrics that agree with human judgments of texture similarity require, in general, extensive subjective tests. The effectiveness and efficiency of such tests depend on a careful analysis of the abilities of human perception and the application requirements. The focus of this paper is on defining performance requirements and testing procedures for objective texture similarity metrics. We identify three operating domains for evaluating the performance of a similarity metric: the ability to retrieve "identical" textures; the top of the similarity scale, where a monotonic relationship between metric values and subjective scores is desired; and the ability to distinguish between perceptually similar and dissimilar textures. Each domain has different performance goals and requires different testing procedures. For the third domain, we propose ViSiProG, a new Visual Similarity by Progressive Grouping procedure for conducting subjective experiments that organizes a texture database into clusters of visually similar images. The grouping is based on visual blending and greatly simplifies labeling image pairs as similar or dissimilar. ViSiProG collects subjective data in an efficient and effective manner, so that a relatively large database of textures can be accommodated. Experimental results and comparisons with structural texture similarity metrics demonstrate both the effectiveness of the proposed subjective testing procedure and the performance of the metrics.

The Effect Of 3D Audio And Other Audio Techniques On Virtual Reality Experience.

Three studies were conducted to examine the effect of audio on people's experience in a virtual world. The first study showed that people could distinguish between mono, stereo, Dolby surround and 3D audio of a wasp. The second study found significant effects for audio techniques on people's self-reported anxiety, presence, and spatial perception. The third study found that adding sound to a visual virtual world had a significant effect on people's experience (including heart rate), while it found no difference in experience between stereo and 3D audio.

The effect of visual context on the identification of ambiguous environmental sounds.

The influence of the specificity of the visual context on the identification of environmental sounds (i.e., product sounds) was investigated. Two different visual context types (i.e., scene and object contexts)-which varied in the specificity of the semantic information-and a control condition (meaningless images) were employed. A contextual priming paradigm was used. Identification accuracy and response times were determined in two context conditions and one control condition. The results suggest that visual context has a positive effect on sound identification. In addition, two types of product sounds (location-specific and event-specific sounds) were observed which exhibited different sensitivities to scene and object contexts. Furthermore, the results suggest that conceptual interactions exist between an object and a context that do not share the same perceptual domain. Therefore, context should be regarded as a network of conceptually associated items in memory.

Memory for product sounds: the effect of sound and label type.

The (mnemonic) interactions between auditory, visual, and the semantic systems have been investigated using structurally complex auditory stimuli (i.e., product sounds). Six types of product sounds (air, alarm, cyclic, impact, liquid, mechanical) that vary in spectral-temporal structure were presented in four label type conditions: self-generated text, text, image, and pictogram. A memory paradigm that incorporated free recall, recognition, and matching tasks was employed. The results for the sound type suggest that the amount of spectral-temporal structure in a sound can be indicative for memory performance. Findings related to label type suggest that 'self' creates a strong bias for the retrieval and the recognition of sounds that were self-labeled; the density and the complexity of the visual information (i.e., pictograms) hinders the memory performance ('visual' overshadowing effect); and image labeling has an additive effect on the recall and matching tasks (dual coding). Thus, the findings suggest that the memory performances for product sounds are task-dependent.