Visual Patterns Research Articles

PSVMA+: Exploring Multi-Granularity Semantic-Visual Adaption for Generalized Zero-Shot Learning.

Generalized zero-shot learning (GZSL) endeavors to identify the unseen categories using knowledge from the seen domain, necessitating the intrinsic interactions between the visual features and attribute semantic features. However, GZSL suffers from insufficient visual-semantic correspondences due to the attribute diversity and instance diversity. Attribute diversity refers to varying semantic granularity in attribute descriptions, ranging from low-level (specific, directly observable) to high-level (abstract, highly generic) characteristics. This diversity challenges the collection of adequate visual cues for attributes under a uni-granularity. Additionally, diverse visual instances corresponding to the same sharing attributes introduce semantic ambiguity, leading to vague visual patterns. To tackle these problems, we propose a multi-granularity progressive semantic-visual mutual adaption (PSVMA+) network, where sufficient visual elements across granularity levels can be gathered to remedy the granularity inconsistency. PSVMA+ explores semantic-visual interactions at different granularity levels, enabling awareness of multi-granularity in both visual and semantic elements. At each granularity level, the dual semantic-visual transformer module (DSVTM) recasts the sharing attributes into instance-centric attributes and aggregates the semantic-related visual regions, thereby learning unambiguous visual features to accommodate various instances. Given the diverse contributions of different granularities, PSVMA+ employs selective cross-granularity learning to leverage knowledge from reliable granularities and adaptively fuses multi-granularity features for comprehensive representations. Experimental results demonstrate that PSVMA+ consistently outperforms state-of-the-art methods.

Analyzing Eye Paths Using Fractals.

Visual patterns reflect the anatomical and cognitive background underlying process governing how we perceive information, influenced by stimulus characteristics and our own visual perception. These patterns are both spatially complex and display self-similarity seen in fractal geometry at different scales, making them challenging to measure using the traditional topological dimensions used in Euclidean geometry.However, methods for measuring eye gaze patterns using fractals have shown success in quantifying geometric complexity, matchability, and implementation into machine learning methods. This success is due to the inherent capabilities that fractals possess when reducing dimensionality using Hilbert curves, measuring temporal complexity using the Higuchi fractal dimension (HFD), and determining geometric complexity using the Minkowski-Bouligand dimension.Understanding the many applications of fractals when measuring and analyzing eye gaze patterns can extend the current growing body of knowledge by identifying markers tied to neurological pathology. Additionally, in future work, fractals can facilitate defining imaging modalities in eye tracking diagnostics by exploiting their capability to acquire multiscale information, including complementary functions, structures, and dynamics.

The Effects of Cognitive Enhancement Drug (Coluracetam) on Visual Perception, Abstract Reasoning, Pattern Recognition, Spatial Orientation, and Analytical Thinking: A Case Study

The Effects of Cognitive Enhancement Drug (Coluracetam) on Visual Perception, Abstract Reasoning, Pattern Recognition, Spatial Orientation, and Analytical Thinking: A Case Study

Integrating Multimodal and Generative AI in Design Research: Enhancing Ethnographic Methods with Data-Driven Analysis

This research examines the application of Multimodal and Generative AI in analyzing interior design trends within the South Korean market. The objective is to enhance traditional ethnographic research methods with AI's quantitative capabilities, offering a more comprehensive understanding of design preferences. This methodology uses AI tools to systematically analyze and categorize design elements from a large dataset of interior design images sourced from a relevant online platform. This process involves extracting key visual trends and patterns using AI-driven semantic analysis and clustering techniques. The study identifies and classifies four dominant design trends, providing a structured overview of current market preferences. By integrating AI into design trend analysis, the study demonstrates a novel approach to understanding consumer preferences, potentially influencing future design decisions. This use of AI in design research broadens the scope of traditional ethnographic studies in design and provides practical insights for designers and industry practitioners, suggesting a shift towards more data-driven design practice.

Comparison Accuracy of CNN and VGG16 in Forest Fire Identification: A Case Study

The current research aims to assess the precision of forest fire detection using CNN and VGG16 models, specifically in the context of fire identification. While both models have demonstrated significant promise in visual pattern recognition, a comprehensive analysis regarding their specific benefits in forest fire identification is still needed. The rationale behind this research stems from the significance of promptly identifying forest fires as a preemptive measure to mitigate their detrimental effects on the environment and society. The employed approach involves the application of transfer learning techniques on a diverse and extensive dataset encompassing different forest fire scenarios. The dataset was used to train both CNN and VGG16 models. The test results indicated that the CNN model achieved a forest fire detection accuracy of 96%, while VGG16 achieved 98% accuracy. The primary objective of this research is to enhance comprehension regarding the merits and demerits of each model in the context of forest fire identification scenarios. While VGG16 exhibits marginally superior performance in identifying forest fires, this discrepancy offers valuable insight into the practical applicability of these two models for fire detection in real-world scenarios. These findings establish a solid basis for the advancement of more dependable and efficient early detection technology in the prevention and management of forest fires in the future. This can be accomplished by capitalizing on the unique capabilities of each model to optimize their performance in practical scenarios.

국제 영화제 포스터를 통한 브랜드 아이덴티티 고찰

Brands that were used to identify products or services are gradually expanding their scope. In addition to companies and product brands, interest in brands is also increasing in the country, individuals, and local governments. International film festivals are also establishing their identity over time, and this appears in various forms. One of them, the poster, is appealing for the film festival by implicitly showing the identity of the film festival. In this study, we analyze what patterns are represented for posters of domestic and overseas international film festivals, focusing on sensory elements, and consider the relationship between sensory elements and brand identity. To this end, we analyzed the meaning, shape, and color elements, focusing on the film festival posters produced since 2000, and found out how their patterns express the film festival identity. As a result of the study, it was confirmed that each film festival showed a clear visual pattern, and through this, style, theme, and overall impression were formed.

Classification of diseases in snake plants using convolutional neural network

snake plant has an important role in human life, as well as in increasing the aesthetic value of the environment. Limited knowledge about diseases in snake plants has a crucial result in improper handling and control when the plant is attacked by disease. Advances in deep learning technology and Convolutional Neural Network (CNN) have presented high opportunities with their advantages in recognizing patterns and features from image data. This research will use a CNN model with VGG-19 architecture to classify diseases in the leaves of the snake plant. It is expected that by using the pre-trained VGG-19 model, the model can recognize complex visual patterns in snake plants. Diseases to be classified include several types of diseases that often attack snake plants such as anthracnose, rust, water soaked lesion, and healthy plants for comparison. The highest value of training accuracy reached a value of 98.08%, validation accuracy of 94.02%, and testing accuracy reached 94%.

Visual Culture: A Study on the Presence of Alphabetical Pattern in ‘Senaman Tua’ by using Motion Capture Technology

Senaman Tua (Malay’s Ancient Exercise) is a therapeutic exercise from Malay culture. This exercise is blend well aesthetically with Malay culture and Islam which give holistic positive effects on the spiritual side of being able to provide, maintain inner peace, patience in the individual and also able to understand their own inner selves. According to the founder, the essence of the holistic positive effect is from the unseen pattern that embedded in the movement. The aim of this study is to visualize the micro-line the embedded in Senaman Tua (Malay’s Ancient Exercise) movement by using Motion Capture Technology. With the permission of Senaman Tua’s founder, Selangkah Guru (A Guru’s movements) is a dynamic exercise movement of Senaman Tua for beginner will be captured by using Motion Capture Technology. There are 33 markers of human anatomy have been captured by using Motion Capture Technology in 1 minute 08 second. The finding show there are presence of alphabetic pattern that generated from the “Selangkah Guru” (A Guru’s Movements). It can be concluded that visual pattern of “Lam Alif” alphabet clearly shows in the mapping of Motion Capture Technology.

A Multimodal Study of How Pronunciation-Induced Communication Breakdowns are Managed During Tandem Interactions

This paper offers quantitative and qualitative findings from the exploration of communication breakdowns in English tandem interactions, by adopting a multimodal perspective. It focuses on the ways in which pronunciation-induced CBs are managed by language peers in a tandem setting. This study shows cases where it was the non-native participant’s output that was the main communicative stumbling block, with a view to reporting on pronunciation-induced breakdowns. More specifically, our analyses target the ways in which CBs are signaled to the interlocutor with different multimodal cues (verbal / vocal / visual). Those pronunciation issues are dealt with in a highly collaborative manner, through multimodal communication strategies, revealing recurrent visual patterns involving different visible body articulators (i.e., the face, the trunk, and the hands) which differ according to participants’ status (native versus non-native).

Screen Printed Electrodes: A Boon for Forensic Chemistry Analysis for Detection of Narcotics and Explosives

Abstract: Screen printing technology is a method of ink-based printing that utilizes a stencil mesh screen to create a visual pattern on a flat surface, such as plastic or silk. This process finds application in the production of screen-printed electrodes, which are utilized in electrochemical cells as part of voltammetric techniques. These electrodes offer advantages in terms of simplicity, portability, and ease of use compared to more intricate techniques. Various voltammetric methods are employed to analyze materials present in trace quantities, including drugs and explosives. By employing screen-printed electrodes in these electro-analytical techniques, it becomes possible to achieve straightforward, cost-effective, sensitive, and rapid detection of analytes with forensic significance. This highlights the significance of screen-printed electrodes in both onsite and laboratory forensic examinations.

Use of real-time near-infrared fluorescence to assess gastric viability in dogs with gastric dilatation volvulus: A case-control study.

To describe near-infrared fluorescence (NIRF) for assessment of gastric viability and describe NIRF's influence on the surgeon's operative strategy in dogs with gastric dilatation and volvulus (GDV). Prospective clinical trial. Twenty dogs with GDV and 20 systemically healthy dogs. Following gastric derotation, the surgeon's subjective assessment of gastric viability was recorded prior to near-infrared imaging. Changes in the surgeon's initial assessment of viability based on the visual pattern of gastric fluorescence was recorded. If nonviable (lack of defined vessels), a partial gastrectomy was performed and submitted for histopathology. The stapled gastrectomy line was imaged. Viable (defined vessels) and nonviable fluorescence intensities were compared with healthy dogs undergoing surgery for nongastrointestinal disease. Subjective assessment diagnosed 17 viable and three nonviable GDVs (2 fundi; 1 cardia). Near-infrared imaging demonstrated nonviable gastric fluorescence in 4 dogs (3 fundi/cardia; 1 fundus). The surgeon's margins for resection were altered in 3/20 dogs. Fluorescence intensity (cardia, fundus, body, pylorus) was lower in GDV viable (30.59%, p = .04; 38.17%, p < .01; 51.18%, p < .01; 44.12%, p= .01) and nonviable (11.00%, p < .01; 4.33%, p < .01; 57.67%, p = .22; 54.33%, p = .72) dogs compared to healthy controls (44.7%, 70.05%, 84.00%, 63.95%). Fundic fluorescence was less in nonviable gastric tissue in comparison with viable gastric tissue (p = .03). Fluorescence of the gastrectomy staple line approximated that of viable tissue. Near-infrared fluorescence can identify histologically confirmed nonviable gastric tissue. These results provide enough evidence to support the implementation of NIRF as an adjunct to gross examination of the gastric wall in dogs with GDV.

Objective Analysis and Clinical Significance of the Spatial Tumor-Infiltrating Lymphocyte Patterns in Non-Small Cell Lung Cancer.

The spatial arrangement of lymphocytes in the tumor bed (e.g., immune infiltrated, immune excluded, immune desert) is expected to reflect distinct immune evasion mechanisms and to associate with immunotherapy outcomes. However, data supporting these associations are scant and limited by the lack of a clear definition for lymphocyte infiltration patterns and the subjective nature of pathology-based approaches. We used multiplexed immunofluorescence to study major tumor-infiltrating lymphocyte (TIL) subsets with single-cell resolution in baseline whole-tissue section samples from NSCLC patients treated with immune checkpoint inhibitors (ICI). The spatial TIL patterns were analyzed using a qualitative pathologist-based approach, and an objective analysis of TIL density ratios in tumor/stromal tissues. The association of spatial patterns with outcomes was studied for different TIL markers. The analysis of CD8+ TIL patterns using qualitative assessment identified prominent limitations including the presence of a broad spectrum of phenotypes within most tumors and limited association with outcomes. The utilization of an objective method to classify NSCLCs showed the existence of at least three subgroups with partial overlap with those defined using visual patterns. Using this strategy, a subset of cases with "immune excluded-like" tumors showed prominently worse outcomes, suggesting reduced sensitivity to ICI; however, these results need to be validated. The analysis for other TIL subsets showed different results, underscoring the relevance of the marker selected for spatial TIL pattern evaluation and opportunities for market integration. Our results identified major challenges associated with the qualitative spatial TIL pattern evaluation. We devised a novel objective strategy to overcome some of these limitations that has strong biomarker potential.

The impact of crossmodal predictions on the neural processing of aesthetic stimuli.

Neuroaesthetic research has focused on neural predictive processes involved in the encounter with art stimuli or the related evaluative judgements, and it has been mainly conducted unimodally. Here, with electroencephalography, magnetoencephalography and an affective priming protocol, we investigated whether and how the neural responses to non-representational aesthetic stimuli are top-down modulated by affective representational (i.e. semantically meaningful) predictions between audition and vision. Also, the neural chronometry of affect processing of these aesthetic stimuli was investigated. We hypothesized that the early affective components of crossmodal aesthetic responses are dependent on the affective and representational predictions formed in another sensory modality resulting in differentiated brain responses, and that audition and vision indicate different processing latencies for affect. The target stimuli were aesthetic visual patterns and musical chords, and they were preceded by a prime from the opposing sensory modality. We found that early auditory-cortex responses to chords were more affected by valence than the corresponding visual-cortex ones. Furthermore, the assessments of visual targets were more facilitated by affective congruency of crossmodal primes than the acoustic targets. These results indicate, first, that the brain uses early affective information for predictively guiding aesthetic responses; second, that an affective transfer of information takes place crossmodally, mainly from audition to vision, impacting the aesthetic assessment. This article is part of the theme issue 'Art, aesthetics and predictive processing: theoretical and empirical perspectives'.

Agonistic behavior of the Baikal seal Pusa sibirica (Gmelin, 1788) on a coastal rookery: aggressive and pacifying patterns

The purpose of the article was to describe the agonistic behavior of the Baikal seal Pusa sibirica at the coastal rookery based on the analysis of video materials obtained during continuous fi lming over several seasons at one of the main coastal rookeries on Dolgiy Island (Ushkany Islands, Lake Baikal). The original ethogram of aggressive and pacifying behavior was compiled for the species, including 20 behavioral patterns. Visual patterns play a predominant role in agonistic relations of seals at coastal rookeries. Due to diverse pacifying patterns, the Baikal seal can gather in relatively numerous haul-outs (tens of individuals) in a limited space of coastal rookeries and coexist quite peacefully without causing signifi cant damage to each other, limited to rare minor injuries. Almost all agonistic interactions of seals on coastal rookeries are caused by competition for territory, a limited resource. The most common is front fl ipper swing and fl ipper hit (imitation and contact). The frequency of its use (48–64%) depends on the density and size of the substrate. Frequency of fi ghts in the water is up to 18%. Other commonly used pacifying patterns include tactile infl uences of varying intensity (up to 55%) and descent into the water or fl ight (8–18%).

Sparsity-guided Discriminative Feature Encoding for Robust Keypoint Detection

Existing handcrafted keypoint detectors typically focus on designing specific local structures manually while ignoring whether they have enough flexibility to explore diverse visual patterns in an image. Despite the advancement of learning-based approaches in the past few years, most of them still rely on the availability of the outputs of handcrafted detectors as a part of training. In fact, such dependence limits their ability to discover various visual information. Recently, semi-handcrafted methods based on sparse coding have emerged as a promising paradigm to alleviate the above issue. However, the visual relationships between feature points have not been considered in the encoding stage, which may weaken the discriminative capability of feature representations for keypoint recognition. To tackle this problem, we propose a novel sparsity-guided discriminative feature representation (SDFR) method that attempts to explore the intrinsic correlations of keypoint candidates, thus ensuring the validity of characterizing distinctive and diverse structural information. Specifically, we first incorporate an affinity constraint into the feature representation objective, which jointly encodes all the patches in an image while highlighting the similarities and differences between them. Meanwhile, a smoother sparsity regularization with the Frobenius norm is leveraged to further preserve the similarity relationships of patch representations. Due to the differentiable property of this sparsity, SDFR is computationally feasible and effective for representing dense patches. Finally, we treat the SDFR model as multiple optimization sub-problems and introduce an iterative solver. During comprehensive evaluations on five challenging benchmarks, the proposed method achieves favorable performances compared with the state of the art in the literature.

Effects of Sleep Deprivation and Hazard Types on the Visual Search Patterns and Hazard Response Times of Taxi Drivers.

The present study attempted to explore the effects of sleep deprivation on the visual search patterns and hazard response times of taxi drivers when they encountered different types of hazards. A two (driver groups: sleep deprivation or control) × two (hazard types: covert hazard or overt hazard) mixed experimental design was employed. A total of 60 drivers were recruited, half of whom were in the sleep-deprived group and half of whom were in the control group. A validated video-based hazard perception test that either contained covert hazards (12 video clips) or overt hazards (12 video clips) filmed from the drivers' perspective was presented to participants. Participants were instructed to click the left mouse button quickly once they detected a potentially dangerous situation that could lead to an accident. Participants' response time and eye movements relative to the hazards were recorded. The sleep-deprived group had a significantly longer response time and took a longer time to first fixate on covert hazards than the control group, while they had a shorter response time to overt hazards than the control group. The first fixation duration of sleep-deprived drivers was longer than that of the control group for overt hazards, while the duration of the first fixation of the two driver groups was similar for covert hazards. Sleep deprivation affects the visual search patterns and response times to hazards, and the adverse effects of sleep deprivation were worse in relation to covert hazards. The findings have some implications for classifying and evaluating high-risk taxi drivers whose hazard perception ability might be affected by insufficient sleep.

What is the visual behaviour and attentional effort of football players in different positions during a real 11v11 game? A pilot study.

Visual perception has been defined as the first step to a football player's decision-making process and it plays an important role in performance in sport. The skill of focussing to prioritize relevant cues has been also considered crucial in sport. This pilot study aims to explore the visual behaviour and attentional effort of three football players (mean age 19 ± 0 years old) in specific-role positions; Right-winger (RW), Centre-Midfielder (CM) and Left-Back (LB), in the five seconds before receiving the ball from their teammate. Twenty-two male football players performed an 11v11 game, where 24 game sequences (trials) from which 166 fixations were recorded and analysed via the Tobii Pro eye-movement registration glasses and software. The gaze behaviour dependent variables were the mean of fixation duration (FD), time to first fixation (TTF), both measured in milliseconds (ms), and the number of fixations (NF) on eight areas of interest (AOIs). AOIs include teammate with and without the ball, opponent without the ball, space around teammate with and without the ball, space around opponent without the ball, ball and undefined. The mean pupil diameter (PD) correlates to the attentional effort and was measured in millimetres (mm). Descriptive statistics showed nonregular search rate data between the participants in FD, TTF, NF on the AOIs. Mean FD on the ball: (CM, 270 ms), (RW, 570 ms), (CM, 380 ms). They also presented differences in the mean PD during play; (CM: 2.90 mm ± 0.26), (RW: 2.74 mm ± 0.30), (LB 2.77mm ± 0.27). Albeit the sample size was small, the findings demonstrated a promising way to measure the on-field perceptual-cognitive abilities of football players according to their specific positions, since different playing roles revealed to present distinctive visual and attentional patterns. This could potentially assist in tailoring players 'visual and focus training.

Learning Dynamic Prototypes for Visual Pattern Debiasing

AbstractDeep learning has achieved great success in academic benchmarks but fails to work effectively in the real world due to the potential dataset bias. The current learning methods are prone to inheriting or even amplifying the bias present in a training dataset and under-represent specific demographic groups. More recently, some dataset debiasing methods have been developed to address the above challenges based on the awareness of protected or sensitive attribute labels. However, the number of protected or sensitive attributes may be considerably large, making it laborious and costly to acquire sufficient manual annotation. To this end, we propose a prototype-based network to dynamically balance the learning of different subgroups for a given dataset. First, an object pattern embedding mechanism is presented to make the network focus on the foreground region. Then we design a prototype learning method to discover and extract the visual patterns from the training data in an unsupervised way. The number of prototypes is dynamic depending on the pattern structure of the feature space. We evaluate the proposed prototype-based network on three widely used polyp segmentation datasets with abundant qualitative and quantitative experiments. Experimental results show that our proposed method outperforms the CNN-based and transformer-based state-of-the-art methods in terms of both effectiveness and fairness metrics. Moreover, extensive ablation studies are conducted to show the effectiveness of each proposed component and various parameter values. Lastly, we analyze how the number of prototypes grows during the training process and visualize the associated subgroups for each learned prototype. The code and data will be released at https://github.com/zijinY/dynamic-prototype-debiasing.

Orthogonal Three‐Wavelength‐Controlled Two‐Stage Supramolecular Self‐Assemblies

AbstractPhotoswitchable supramolecular self‐assemblies are promising candidates for intelligent recognition because of their noninvasive and spatial precision. However, the manipulation of the self‐assembly and disintegration of multicomponent mixtures with light of different wavelengths remain difficult. Here, an orthogonal three‐wavelength‐controlled two‐stage supramolecular self‐assembly consisting of two photoswitches, diarylethene, and protonated merocyanine is designed, mediated by supramolecular macrocyclic host. Owing to the reversible photo‐transformation of the switches, an optimized approach including three wavelengths irradiation and a thermal relaxation process is developed without mutual interference, realizing interconversion among four self‐assembled states. These self‐assemblies are photodeformable with reversibility, which can be operated in a cellular environment. Taking advantage of the photochromism in both absorption and fluorescence of these self‐assemblies in different behaviors, a visual pattern that can simultaneously display multiple hidden information under dual‐signal emission is created. This work signifies a promising strategy in improving the ability of information encryption and anti‐counterfeiting.

Experimental, Modal, and Harmonic Response Analysis of a Chladni Plate at Ultrasonic Frequencies

Mode shapes and natural frequencies of mechanical structures can be determined with the Chladni approach. Visual patterns are generated on the Chladni plate if it is exposed to vibrations. These patterns are known as Chladni patterns/figures. Mechanical vibrators are used to excite the plates at particular frequencies for pattern generation, but based on the making and application, their range is limited up to 10 kHz. Piezoelectric transducers are a special kind of transducers that are capable of generating ultrasonic frequencies and if are attached to a plate, intricate visual patterns can be generated, based on the material properties and the shape of the plate. The current research focuses on experimenting with and simulating Chladni figures in the ultrasonic frequency range. The simulations were performed in ANSYS Workbench as a validation of the experimental work.