Sight Research Articles

Machine visual perception from sim-to-real transfer learning for autonomous docking maneuvers

AbstractThis paper presents a comprehensive approach to enhancing autonomous docking maneuvers through machine visual perception and sim-to-real transfer learning. By leveraging relative vectoring techniques, we aim to replicate the human ability to execute precise docking operations. Our study focuses on autonomous aerial refueling as a use case, demonstrating significant advancements in relative navigation and object detection. We introduce a novel method for aligning digital twins using fiducial targets and motion capture data, which facilitates accurate pose estimation from real-world imagery. Additionally, we develop cost-efficient annotation automation techniques for generating high-quality You Only Look Once training data. Experimental results indicate that our transfer learning methodologies enable accurate and reliable relative vectoring in real-world conditions, achieving error margins of less than 3 cm at contact (when vehicles are approximately 4 m from the camera) and maintaining performance at over 56 fps. The research findings underscore the potential of augmented reality and scene augmentation in improving model generalization and performance, bridging the gap between simulation and real-world applications. This work lays the groundwork for deploying autonomous docking systems in complex and dynamic environments, minimizing human intervention and enhancing operational efficiency.

Experimental study of hue cancellation and its mechanism exploration

To research the distinctions between dichoptic color vision and naked-eye color vision in the human vision system, this study carried out psychophysical experiments to explore hue cancellation phenomena in both dichoptic and naked-eye visual conditions. The experimental results show that dichoptic hue cancellation (DHC) not only occurs in the red and green (R-G), yellow and blue (Y-B), and yellow-green and purple (YG-P) color directions in CIELAB space but also requires color desaturation. In naked-eye vision, a specific color for each color is capable of inducing hue cancellation. The experimental results imply that hue cancellation in naked-eye vision may be a complementary phenomenon. The process of viewing two hues to produce a third hue should be completed in the retina.

Feature Weighted Cycle Generative Adversarial Network with Facial Landmark Recognition and Perceptual Color Distance for Enhanced Face Animation Generation

We propose an anime style transfer model to generate anime faces from human face images. We improve the model by modifying the normalization function to obtain more feature information. To make the face feature position of the anime face similar to the human face, we propose facial landmark loss to calculate the error between the generated image and the real human face image. To avoid obvious color deviation in the generated images, we introduced perceptual color loss into the loss function. In addition, due to the lack of reasonable metrics to evaluate the quality of the animated images, we propose the use of Fréchet anime inception distance to calculate the distance between the distribution of the generated animated images and the real animated images in high-dimensional space, so as to understand the quality of the generated animated images. In the user survey, up to 74.46% of users think that the image produced by the proposed method is the best compared with other models. Also, the proposed method reaches a score of 126.05 for Fréchet anime inception distance. Our model performs the best in both user studies and FAID, showing that we have achieved better performance in human visual perception and model distribution. According to the experimental results and user feedback, our proposed method can generate results with better quality compared to existing methods.

Visible-Light-Stimulated Optoelectronic Neuromorphic Transistor Based on Indium-Gallium-Zinc Oxide via Bi2Te3 Light Absorption Layer.

To emulate a visual perception system, a bismuth telluride (Bi2Te3)/indium-gallium-zinc oxide (IGZO) heterostructure is introduced for optoelectronic neuromorphic transistors (ONTs). Amorphous IGZO is applied as a channel layer to exhibit low off-current, high mobility, and persistent photoconductivity, enabling light-stimulated neuromorphic characteristics. The atomic ratio of In/Ga/Zn was 9.4:9.8:5.2. For a light absorption layer, Bi2Te3 is applied due to a small bandgap, high photoresponse, and carrier concentration. However, conventional optoelectronic devices using Bi2Te3 exhibit insufficient performance owing to their excessive conductivity. To resolve the constraint, the oxidation of Bi2Te3 is performed to suppress its electrical conductivity. Finally, the IGZO ONT with a Bi2Te3 layer exhibits optoelectronic characteristics under visible-light irradiation. Under red-light irradiation having a light intensity of 5 mW mm-2, it exhibits enhanced optoelectronic characteristics including photoresponsivity, photosensitivity, and detectivity from 19.6 to 3.46 × 102 A/W, 4.95 to 1.46 × 108, and 1.45 × 107 to 2.13 × 1012 Jones compared to that without a Bi2Te3 layer, respectively. To confirm reproducibility and uniformity, we fabricated and compared the optoelectronic characteristics of 5 samples from batch to batch. Regarding optoelectronic neuromorphic characteristics, the ONT exhibits short- and long-term memory and a highly linear relationship between peak synaptic current and pulse number under red-light pulses. The paired-pulse facilitation index was 208%. To confirm the applicability of artificial visual perception, a 6 × 6 ONT array demonstrates long-term memory characteristic successfully after red-light irradiation with a Y-shaped pattern. In view of stability and reliability, we conducted three potentiation processes and compared the peak synaptic currents of each potentiation process.

The PAUL Glaucoma Implant in the management of uveitic glaucoma-3-year follow-up.

Evaluate the mid-term outcomes of the PAUL Glaucoma Implant (PGI) in the management of uveitic glaucoma. This was a single-centre, multi-surgeon, retrospective analysis of 50 consecutive cases of PGI for uveitic glaucoma performed between April 2019 and August 2021. Primary outcomes include: complete and qualified success (IOP ≥5 mmHg to ≤21 mmHg with ≥20% IOP reduction) or failure (IOP exceeding the success criteria, additional glaucoma procedures, no perception of light vision). Secondary outcomes included: visual acuity, IOP, medications, complications and intraluminal stent removal. We included 50 eyes of 41 patients. Mean age was 45.8 ± 19.8 years (range 6-81 years) in this heterogenous and complex cohort. Mean pre-op IOP was 30.6 ± 9.8 mmHg on 3.9 ± 0.9 medications. In total, 62% of patients were on acetazolamide, and 64% required systemic immunosuppression. At final follow-up (mean: 35.8 ± 9.8 months, range 5-58 months), IOP and medications were significantly reduced (12.2 ± 4.4 mmHg, requiring 1.1 ± 1.3 medications, p < 0.0001). Resulting in 48% complete and 92% qualified success rates. Failure occurred in 8% of cases, 6% due to hypertension but only one case of prolonged hypotony (2%). To date, this study represents the first publication looking specifically at the efficacy and safety of the PGI in the management of complex uveitic glaucoma. With an average follow-up of 3 years, it shows high levels of complete and qualified success with few complications.

Pengaruh Kualitas Produk, Citra merek dan Persepsi Harga Terhadap Kepuasan Pelanggan Produk Smartphone Xiaomi: Studi pada Konsumen Produk Xiaomi di Kabupaten Jepara

This study was conducted to analyze the effect of product quality, brand image and price perception on customer satisfaction of Xiaomi HP in Jepara Regency. The sample in this study amounted to 100 respondents with the sampling method using Non probability Sampling technique and using google form as a medium with a questionnaire method. This research analysis uses multiple linear regression analysis. The results of the research instrument test show that all indicators are valid and all variables are reliable. Based on multiple regression analysis obtained results: (1) product quality has no significant effect on customer satisfaction (2) brand image has a positive and significant effect on customer satisfaction (3) price perception has a positive and significant effect on customer satisfaction.

Correlation reconstruction mechanism based on dual wavelength imaging and neural network

In a natural light-field imaging scene, lighting is usually performed using a mixed light field of multiple wavelengths. To better meet the requirements of human vision, this study proposes a correlation reconstruction method based on dual-wavelength imaging and a neural network (compression sensing correlation fusion-reconstruction by dual-wavelength imaging and auto-encoder neural network, CSCFR-DWI-AENN). Two different wavelengths of light are mixed through an optical multiplexer unit (OMU) to form a dual-wavelength illumination light field, and the light intensity value reflected by the object is received by the detector. The object information was reconstructed using the compressed sensing ghost imaging algorithm, and the distribution of the illumination field was taken as prior information. Two single-wavelength illumination information and the detection value are used for compressed sensing reconstruction, and the high-quality reconstructed image of the target to be measured is obtained by combining the noise reduction advantage of the autoencoder neural network. Then the target information is fused by the New Sum of Modified Laplacian (NSML) algorithm. Two single-wavelength reconstruction, dual-wavelength direct reconstruction and the proposed reconstruction were compared respectively, and non-reference image quality evaluation functions such as EN, MI, EAV and SF were used to process and analyze the reconstruction effect. It is proved that this algorithm can reconstruct the object image comprehensively, reproduce the complete detail information, and achieve a more accurate, more comprehensive and more reliable description of the same object. It provides a theoretical basis for multi-wavelength imaging technology and has a good application prospect.

Industry Image Perceptions and Organizational Attractiveness: Results of an International Survey

ABSTRACTExtending the research on corporate brand image and recruitment, this study investigates the influence of industry image on organizational attractiveness in a cross–national context. Drawing from signaling theory and the application of the instrumental‐symbolic image framework, we apply an experimental vignette design in the renewable energy industry (REI) and the oil and gas exploration and production industry (OGI) with potential job applicants in France, Germany, and the United Kingdom. We conceptualize national context as collective signaling environments. Results from 550 respondents indicate that the REI and OGI differ in most of the instrumental and symbolic image dimensions. The instrumental image dimension of pay and the three symbolic image dimensions of sincerity, innovativeness, and prestige predict organizational attractiveness. Moreover, sincerity, innovativeness, and prestige carry a mediating effect in the link between industry affiliation and organizational attractiveness. Findings also demonstrate that image perceptions of the REI and OGI vary across countries such that the already more favorable image perceptions regarding the REI are for the most part even more pronounced in France. Implications for theory and tailored recommendations for practice are provided.

Neutron-Image Super-Resolution based on Convolutional Neural Networks and Novel Parallel Transformer

Enhancing the resolution of neutron imaging can be achieved by upgrading the hardware components of the neutron imaging system, including employing more potent neutron source systems, utilizing high-resolution detectors, and implementing superior collimation techniques. However, such improvements are frequently constrained by challenges such as high cost and limited flexibility. In this paper, a flexible and efficient single-neutron-image super-resolution network is proposed for neutron image super-resolution. We propose a novel multi-scale feature extraction network, which can extract features deeply by fusing features of different types and levels. In order to learn more context information, a novel parallel Transformer is proposed to overcome the long dependency problem of sequences. A high-frequency information enhancement network is proposed to enhance the high-frequency information of neutron image in order to address the blur problem. Quantitative and qualitative analyses of benchmark datasets and real neutron images show that the proposed network performs better on Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM) than bicubic interpolation and other methods, and reconstructs high-resolution neutron images that are more consistent with human visual perception, highlighting its application potential in neutron images quality enhancement.

Orthography Mediates the Unconscious Semantic Processing of Chinese Characters.

Whether conscious awareness is critical for detecting semantic information in individual words remains debated. We hypothesize that this issue is specific to language type and that orthography serves as a mediator between semantics and conscious awareness. Using a priming-based paradigm called breaking continuous flash suppression (b-CFS) with a full factorial design, this study investigated whether orthographic similarity and semantic relatedness between the prime and target could modulate conscious access to Chinese single characters. Statistics indicated that semantic relatedness did not facilitate the character's breakthrough from CFS unless combined with orthographic similarity, although orthographic similarity alone had an inhibitory effect. The results were discussed in light of the "unconscious binding" hypothesis and object-updating theory in visual perception as well as the unique sub-morphemic construction in psycholinguistics. Our findings demonstrate the possibility of orthography-based semantic processing occurring outside of conscious awareness and suggest that the multilevel interaction activation model may be applicable to the early recognition of Chinese characters.

Simple actions modulate context-dependent visual size perception at late processing stages

A simple button press towards a prime stimulus enhances subsequent visual search for objects that match the prime. The present study investigated whether this action effect is a general phenomenon across different task domains, and the underlying neural mechanisms. The action effect was measured in an unspeeded size-matching task, with the presentation of the central target and the surrounding inducers of the Ebbinghaus illusion together to one eye or separately to each eye, and when repetitive TMS was applied over right primary motor cortex (M1). The results showed that a prior key-press significantly reduced the illusion effect compared to passive viewing. Notably, the action effect persisted with dichoptic presentation of the Ebbinghaus configuration, but disappeared with the right M1 disruption. These results suggest that action guides visual perception to influence human behavior, which mainly affects the late visual processing stage and probably relies on feedback projections from the motor cortex.

Quantitative Characterization of Highway Landscape Space Visual Perception Based on Deep Learning

Quantitative Characterization of Highway Landscape Space Visual Perception Based on Deep Learning

Q-BENCH: A Benchmark for Multi-modal Foundation Models on Low-level Vision from Single Images to Pairs.

The rapid development of Multi-modality Large Language Models (MLLMs) has navigated a paradigm shift in computer vision, moving towards versatile foundational models. However, evaluating MLLMs in low-level visual perception and understanding remains a yet-to-explore domain. To this end, we design benchmark settings to emulate human language responses related to low-level vision: the low-level visual perception (A1) via visual question answering related to low-level attributes (e.g. clarity, lighting); and the low-level visual description (A2), on evaluating MLLMs for low-level text descriptions. Furthermore, given that pairwise comparison can better avoid ambiguity of responses and has been adopted by many human experiments, we further extend the low-level perception-related questionanswering and description evaluations of MLLMs from single images to image pairs. Specifically, for perception (A1), we carry out the LLVisionQA+ dataset, comprising 2,990 single images and 1,999 image pairs each accompanied by an open-ended question about its low-level features; for description (A2), we propose the LLDescribe+ dataset, evaluating MLLMs for low-level descriptions on 499 single images and 450 pairs. Additionally, we evaluate MLLMs on assessment (A3) ability, i.e. predicting score, by employing a softmax-based approach to enable all MLLMs to generate quantifiable quality ratings, tested against human opinions in 7 image quality assessment (IQA) datasets. With 24 MLLMs under evaluation, we demonstrate that several MLLMs have decent low-level visual competencies on single images, but only GPT-4V exhibits higher accuracy on pairwise comparisons than single image evaluations (like humans). We hope that our benchmark will motivate further research into uncovering and enhancing these nascent capabilities of MLLMs. Datasets will be available at https://github.com/Q-Future/Q-Bench.

Visual perception enhancement fall detection algorithm based on vision transformer

Visual perception enhancement fall detection algorithm based on vision transformer

Icon similarity model based on cognition and deep learning

Human-computer cooperation guided by natural interaction, intelligent interaction, and human–computer integration is gradually becoming a new trend in human–computer interfaces. An icon is an indispensable pictographic symbol in an interface that can convey pivotal semantics between humans and computers. Research on similar icons’ cognition in humans and the discrimination of computers can reduce misunderstandings and facilitate transparent cooperation. Therefore, this research focuses on images of icons, extracted contours, and four features, including the curvature, proportion, orientation, and line of the contour, step by step. By manipulating the feature value change to obtain 360 similar icons, a cognitive experiment was conducted with 25 participants to explore the boundary values of the feature dimensions that cause different levels of similarity. Its boundary values were applied to deep learning to train a discrimination algorithm model that included 1500 similar icons. This dataset was used to train a Siamese neural network using a 16-layer network branch of a visual geometry group. The training process used stochastic gradient descent. This method of combining human cognition and deep learning technology is meaningful for establishing a consensus on icon semantics, including content and emotions, by outputting similarity levels and values. Taking icon similarity discrimination as an example, this study explored the analysis and simulation methods of computer vision for human visual cognition. The accuracy evaluated is 90.82%. The precision was evaluated as 90% for high, 80.65% for medium, and 97.30% for low. Recall was evaluated as 100% for high, 89.29% for medium, and 83.72% for low. It has been verified that it can compensate for fuzzy cognition in humans and enable computers to cooperate efficiently.

Impact of Body Image Perception on Behavioral Outcomes in Chinese Adolescent and Young Adult Survivors of Sarcoma.

To identify the prevalence and predictive factors of body image dissatisfaction among Chinese adolescent and young adult (AYA) survivors of sarcoma and to evaluate its associations with behavioral outcomes. In total, 116 AYA survivors (response rate: 88%; 48.3% female; mean age 28.2 years, SD = 8.2 years) of osteosarcoma (49.1%) or soft-tissue sarcoma (50.9%) were recruited from an academic hospital. The survivors self-reported their perceived body image using the Body Image Scale. Behavioral outcomes were assessed using DSM-oriented scales of the ASEBA Adult Self-Report checklist. Multivariable linear regression was conducted to identify predictors of body image perception and to investigate the association between body image dissatisfaction and behavioral outcomes (adjusted for clinically relevant variables and corrected for false discovery rate). Mediation analyses were performed to examine the mediating effects of body image perception between clinical, treatment, or socioenvironmental risk factors and behavioral outcomes. At 15 years post-cancer diagnosis, one-third of the cohort (39.7%) reported dissatisfaction with their body image. The significant risk factors associated with body image dissatisfaction were being female (standardized coefficient estimate [Est] = 0.21, SE = 0.10; p = 0.047), surgery treatment (Est = 0.19, SE = 0.09; p = 0.046), and worse family functioning (Est = 0.27, SE = 0.10, p = 0.013). Body image dissatisfaction was associated with symptoms of depression (Est = 0.30, SE = 0.10; p = 0.005) and avoidant personality problems (Est = 0.37, SE = 0.11; p < 0.001). Negative body image perception significantly mediated the association between poor family functioning and avoidant personality problems (proportion-mediated 26.3%, p = 0.038). Body image dissatisfaction was associated with more significant internalizing symptoms of depression, anxiety, and avoidant personality. A negative body image perception may mediate the association between poor family functioning and psychological distress among adult survivors. The provision of psychosocial intervention early during the cancer care continuum may mitigate the negative impact of body image distress in Chinese AYA survivors.

Distortion of body image perception in the Prader-Willi syndrome: Relationship with the perceptual reasoning index

IntroductionSelf-perception of body image has been scarcely evaluated in people with Prader-Willi syndrome (PWS), who, in addition to intellectual disability, are often obese. Therefore, we explored whether people with PWS can accurately identify their true image and how this self-perception is impacted by their neuropsychological profile. MethodologyThis observational study included patients with PWS with regular attendance to transdisciplinary treatment at a center specialized in the management of rare diseases. All patients were evaluated with the Stunkard scale (including silhouettes ranging from extremely skinny to extremely obese) and the WISC-IV and WAIS-III questionnaires, specifically the perceptual reasoning index (PRI). ResultsAmong the 21 participants, 62% misperceived their body image, most underestimating their body dimensions (actual BMI 28.0±8.3kg/m2 vs self-perceived BMI 23.2±4.7kg/m2, p=0.03). While BMI differences between accurate and inaccurate body image perception were nonsignificant (accurate 26.6±8.8kg/m2 vs inaccurate 28.9±8.1kg/m2, p=0.56), individuals with accurate perception showed both higher PRI scores (accurate 67.6±8.2 vs inaccurate 60.2±7.2, p=0.043) and, to a lesser extent, intelligence quotients (IQ) (accurate 60.0±4.9 vs inaccurate 55.0±7.5, p=0.079). ConclusionIn this study, we identified distortion of body image perception as a very common finding among PWS patients, in most cases as underestimation, and influenced by the neuropsychological profile.

Effect of Virtual Reality-Based Upper Extremity Training on Upper Extremity Function, Visual Perception, and Brain Activ ity in a Child with Spastic Cerebral Palsy: A Case Report

Effect of Virtual Reality-Based Upper Extremity Training on Upper Extremity Function, Visual Perception, and Brain Activ ity in a Child with Spastic Cerebral Palsy: A Case Report

Research on Relationship Between Image Perception, Place Attachment and Tourists' Pro-environment Behavior

Ice and snow tourism, as a special way of tourism, is booming with the continuous development and utilization of ice and snow tourism resources in the world. China is rich in ice and snow tourism resources. In recent years, the ice and snow tourism industry has developed rapidly. However, problems such as ecological destruction and over-exploitation of resources have become increasingly prominent. In this context, pro-environment behavior has become an important means to alleviate the environmental pressure of ice and snow tourism destinations; At the same time, research on the relationship between the perceived image of ice and snow tourism destination, place attachment and tourists' pro-environment behavior has important theoretical value and practical significance for promoting the sustainable development of ice and snow tourism, enhancing tourists' environmental awareness and responsibility, and improving the image and attractiveness of ice and snow tourism destination. Based on the cognitive affective theory, place attachment, etc., and on the basis of literature research, this thesis proposes several hypotheses, constructs conceptual models, develops or cites corresponding measurement scales for the relationship among the four variables of cognitive image, emotional image, place attachment and pro-environment behavior, and takes Chongli Ski Resort as a case for empirical research. Using questionnaire survey data, the structural equation model of 264 valid questionnaires was analyzed to suggest the influence mechanism of cognitive image, emotional image and place attachment on tourists' pro-environment behavior. The findings are as follows: (1) Cognitive image has significant positive effects on affective image, place attachment and pro-environment behavior, and affective image and place attachment can mediate this process; (2) Emotional image has a significant positive impact on place attachment and pro-environment behavior, and place attachment can be used as a mediator to have a positive impact on them; (3) Place attachment has a significant positive effect on pro-environment behavior. As an emotional bond, it connects the emotional connection between tourists and tourist destinations. Accordingly, this study also puts forward the strategy to promote the tourists' pro-environment behavior in Chongli ski Resort in order to promote the sustainable development of ice and snow tourism. Including: (1) to enhance the attractiveness of ice and snow tourism; (2) Enhance the emotional connection between places and tourists; (3) Strengthen environmental awareness education. It is hoped that this study will have some reference value for the improvement of pro-environment behavior of tourists in other similar ice and snow tourism destinations.

Autonomous Street Lighting System based on Solar Energy and LEDs

This paper introduces a solar-powered autonomous street lighting system designed primarily for areas without access to the grid, such as rural roads and intersections. It utilizes solar panels as the primary energy source, with batteries serving as backup, and employs light-emitting diodes (LEDs) for illumination. The proposed system is optimized for efficiency by employing direct current (DC) power at all stages. To replace a 70W high-pressure sodium (HPS) lamp, an LED fixture design is discussed, considering human scotopic vision. Experimental outcomes for the LED driver and battery charger are provided, demonstrating features such as maximum power point tracking (MPPT), constant current, and constant voltage modes. The system ensures MPPT under varying solar irradiance and temperature conditions by analyzing the battery charger input impedance.